WO2025111647A1

WO2025111647A1 - Method of assessing lung health

Info

Publication number: WO2025111647A1
Application number: PCT/AU2024/051266
Authority: WO
Inventors: Andreas Fouras
Original assignee: 4DMedical Ltd
Current assignee: 4DMedical Ltd
Priority date: 2023-11-27
Filing date: 2024-11-27
Publication date: 2025-06-05
Anticipated expiration: 2026-05-27

Abstract

A method of identifying changes in lung health using 3-dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath; calculating a relative regional volume of a region of the lung for each of the two 3-dimensional image datasets; comparing the relative regional volumes calculated at each breath to assess a change in relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

Description

METHOD OF ASSESSING LUNG HEALTH

Field of Invention

[0001] The present invention relates to method and apparatus for identifying changes in lung health by assessing changes in the lungs over time.

Related Applications

[0002] This application claims priority from Australian Patent Application 2023903814, filed 27 November 2023 entitled “Method of Assessing Lung Health”, which is hereby incorporated herein by reference in its entirety.

Background

[0003] Illness and disease can affect the function and form of moving body tissue in human organs, such as the heart or lung, and can result in deterioration or damage to the organ. Imaging of human organs can provide great benefits to medical research and clinicians in the study, diagnosis and treatment of illnesses and diseases affecting those organs. Accurate imaging of a particular organ, such as a heart or lung, assists clinicians to correctly diagnose and treat problems.

[0004] 3-dimensional (3D) representations of body tissue obtained from X-ray imaging apparatus, including 3D CT (computerised tomography) scanners, offer non-invasive techniques of providing in vivo internal detailed images. CT scanners capture X-ray images taken at different angles around a body and use computer processing to combine the images to create cross-sectional images (axial slices) through the body. CT scanner technology is well developed and commonly used by clinicians to obtain detailed internal images of patients to help diagnose health problems.

[0005] Some systems allow a time component to be added by recording time series of sequential images, providing the benefit of viewing the images in 4 dimensions, namely 3D with the addition of time. These 3-dimensional and 4-dimensional imaging techniques can be used to obtain accurate images and representations of human organs, including the heart and lungs.

[0006] Patients experiencing illness or disease affecting the function or form of their body organs may be placed on a treatment plan to prevent further deterioration or damage, or to reverse the effects of existing deterioration or damage. However, monitoring changes in organ function resulting from these treatments is complex and the ability to assess the effectiveness of these treatments is limited. Even in its simplest set of possibilities: 1) diseased tissue of an organ can be worsening in function due to disease progression, can be stable, or can be improving in function due to treatment, and 2) healthy tissue could be remaining healthy, becoming effected by disease, or could be damaged by ‘off target’ effects of the treatment.

[0007] It is desirable for clinicians to track the effects of treatment on organ function or form by comparing images of the organ before and after treatment. However, despite state of the art imaging techniques providing accurate images and representations of human organs, it can be problematic for clinicians to make reliable comparisons between images captured during different imaging sessions. One problem, in particular for lung imaging, is that it is unlikely that the breath-hold CT images are captured at exactly the same respiration volume in the different imaging sessions, and so the region of interest (i.e. lung tissue of interest) may be a different volume and/or a different shape in each CT image, resulting in clinicians being unable to meaningfully directly compare the CT images.

Summary of the Invention

[0008] Embodiments of the present invention provide a method and apparatus for assessing lung health by assessing the distribution of air in the lungs over two images taken at different times.

[0009] Embodiments of the present invention provide a method and apparatus for assessing lung health by identifying changes in the distribution of volume of the lungs at different breaths. Embodiments calculate the distribution of volume of the lung at a first breath and the distribution of volume of the lung at a second breath and compare the distribution of volume. A change in the distribution of volume of the lung may be indicative of a change in lung health, for example the onset of a lung condition. A change in distribution of volume is a good approximation to the distribution of air within the lung and so by comparing the distribution of volume between images from different scans an assessment can be made of how distribution of air within the lung has changed over time between scans. This change in distribution of air can provide important information about change in lung function and lung health.

[0010] Embodiments retrieve volume data for lungs for at least two different breaths. Embodiments select a region of the lung for comparison. The region may be defined by lung tissue and referred to as a lung tissue region. The relative volume of the region of the lung is calculated with respect to a comparative larger part of the lung, for example the left lung, the right lung, or the entire lungs (i.e. both the left and the right lungs). The comparative larger part of the lung may be a selected portion of the lung. The relative volume of the region is calculated by calculating the volume of the region and calculating the volume of the comparative larger part of the lung. The relative volume of the region expresses the volume of the region with respect to the volume of the comparative larger part. This may be calculated by dividing the volume of the region by the volume of the comparative larger part of the lung. The relative volume is unitless and may be expressed as a fraction, or percentage of the volume of the comparative larger part of the lung. The relative volume of the region is calculated for different breaths. The results are compared to identify any change in distribution of volume of the lung. A change in the relative volume of a particular region of the lung is an indication that the distribution of volume of air in the lung is different at the different breaths.

[0011] Embodiments provide a method of identifying changes in lung health using 3- dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath; calculating a relative regional volume of a region of the lung for each of the two 3-dimensional image datasets; comparing the relative regional volumes calculated at each breath to assess a change in relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

[0012] The step of calculating a relative regional volume of a region of the lung may be calculated with respect to a comparative larger part of the lung. The comparative larger part of the lung may be one of: the left lung; the right lung; or, the combination of the left lung and the right lung. The relative regional volume of the region of the lung may be expressed as a proportion, fraction or a percentage of the volume of the comparative larger part of the lung.

[0013] Further embodiments provide a method of identifying changes in lung health using 3- dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath of a lung; calculating a relative regional distribution of air in the lung for each of the two 3-dimensional image datasets; comparing the relative regional distributions of air in the lung calculated at each breath to assess a change in relative regional distribution of air in the lung between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

[0014] The step of comparing the relative regional distribution of air in the lung comprises the steps of determining the total volume of air in the lung for each breath, normalizing the image datasets to a common total volume of air, and comparing the normalized image datasets by comparing the volume of air in equivalent regions of the normalized image datasets. The step of calculating a relative regional distribution of air in the lung for each of the two 3-dimensional image datasets may comprise the step of dividing the 3-dimensional image datasets into at least two lung regions and calculating the regional distribution of air in the lung for each of the at least two regions.

[0015] The region of the lung may be one of the superior lobe of the left lung, the inferior lobe of the left lung, the superior lobe of the right lung, the middle lobe of the right lung and the inferior lobe of the right lung.

[0016] The lungs may be divided into one of: more than five regions, at least 5 regions; at least 6 regions; at least 10 regions; at least 15 regions; at least 18 regions; at least 19 regions; at least 20 regions; at least 25 regions; at least 40 regions; at least 50 regions; or, at least 100 regions.

[0017] The region may be defined by at least one voxel in a 3-dimensional image of the lung.

[0018] Each of the 3-dimensional images of the lungs may be acquired using computerized tomography (CT) technique. At least one of the images may be acquired at an inspiration breath hold.

[0019] The method may be performed without requiring a visual image, whether digital, physical or otherwise, to be produced.

[0020] Further embodiments provide a method of identifying changes in lung health using 3- dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquiring two 3-dimensional image datasets of a lung, each image dataset being acquired during a different breath of a lung; comparing the image datasets to assess a change in relative regional distribution of air in the lungs between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

[0021] Further embodiments provide a system for identifying changes in lung health using 3- dimensional image datasets, including: an image acquisition module, configured to acquire two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath; a processor configured to calculate a relative regional volume of a region of the lung for each of the two 3-dimensional image datasets; and compare the relative regional volumes calculated at each breath to assess a change in relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

[0022] Further embodiments provide a non-transitory computer readable storage medium having a computer program stored therein, that when executed by a processor of a computer, causes the computer to execute steps directed to identifying changes in lung health using 3- dimensional image datasets, including: acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath; calculating a relative regional volume of a region of the lung for each of the two 3-dimensional image datasets; comparing the relative regional volumes calculated at each breath to assess a change in relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

[0023] Further embodiments provide a method of identifying changes in lung health using 3- dimensional images, including acquiring two 3-dimensional image datasets of a lung, each 3- dimensional image dataset being acquired during a different breath of a lung; normalizing the two three-dimensional image datasets to produce two normalized 3-dimensional image datasets, comparing the two normalized 3-dimensional image datasets to assess a change in relative regional distribution of air in the lung between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

[0024] Further embodiments provide a method of identifying changes in lung health using two 3-dimensional image datasets, including: acquiring a first 3-dimensional image dataset of a lung and a second 3-dimensional image dataset of the lung, the second 3-dimensional image dataset being acquired in a subsequent breath to the first 3-dimensional image dataset; defining a lung tissue region in one of the first and second 3-dimensional image datasets, the lung tissue region consisting of one or more voxels in the 3-dimensional image dataset; mapping the lung tissue region to the corresponding lung tissue in the other 3-dimensional image dataset to define a corresponding lung tissue region; normalizing the volume of the lung tissue region to the volume of the 3-dimensional image dataset it was derived from, and normalizing the volume of the corresponding lung tissue region to the volume of the 3-dimensional image dataset it was derived from, to create two relative regional volumes; comparing the two relative regional volumes to calculate a change in the relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

[0025] Further embodiments provide a method of identifying changes in lung health using two 3-dimensional image datasets, including: acquiring a first 3-dimensional image dataset of a lung in a first breath, and a second 3-dimensional image dataset of the lung in a second breath, the first breath and the second breath being different breaths; defining a lung tissue region in the first 3-dimensional image dataset, the lung tissue region consisting of one or more voxels in the 3-dimensional image dataset; mapping the lung tissue region in the first 3-dimensional image dataset to the corresponding lung tissue in the second 3-dimensional image dataset to define a corresponding lung tissue region in the second 3-dimensional image dataset; normalizing the volume of the lung tissue region in the first 3-dimensional image dataset to the volume of the first 3-dimensional image dataset to create a first regional volume distribution measurement; normalizing the volume of the corresponding lung tissue region in the second 3-dimensional image dataset to the volume of the second 3-dimensional image dataset to create a second regional volume distribution measurement; comparing the first regional volume distribution measurement to the second regional volume distribution measurement to calculate a change in the regional volume distribution measurement between the two breaths, wherein the change in the regional volume distribution measurement is indicative of a change in lung health.

[0026] Further embodiments provide a method of identifying changes in lung health using 3- dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath of a lung; calculating a relative regional distribution of air in the lung for each of the two 3-dimensional image datasets; comparing the relative regional distributions of air in the lung calculated at each breath to assess a change in relative regional distribution of air in the lung between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

[0027] The step of calculating a relative regional distribution of air in the lung may comprise the steps of calculating the total volume of air in the lung and for each of a plurality of regions of the lung calculating the proportion of the total air in the lung in each of the plurality of regions of the lung. The proportion may be at least one of a fraction or a percentage.

[0028] The step of calculating a relative regional distribution of air in the lung may comprise the step of calculating the total volume of air in the lung and for each of a plurality of regions of the lung calculating the volume of air in each of the plurality of regions of the lung. The step of comparing the relative regional distribution of air in the lung may comprise the steps of determining the total volume of air in the lung for each breath, normalizing the image datasets to a common total volume of air, and comparing the normalized image datasets by comparing the volume of air in equivalent regions of the normalized image datasets. At least one of the 3- dimensional image datasets of a lung may be modified during the step of normalizing the images.

[0029] The method may comprise the further step of deforming at least one of the 3- dimensional image datasets to bring the 3-dimensional image datasets into the same coordinate system. The step of deforming may be performed using at least one of: deformable image registration (DIR) algorithms, or particle image velocimetry (PIV).

[0030] The step of calculating a relative regional distribution of air in the lung for each of the two 3-dimensional image datasets may comprise the step of dividing the 3-dimensional image datasets into at least two lung regions and calculating the regional distribution of air in the lung for each of the at least two regions. The at least two lung regions may comprise five lobes of the lungs, being the superior lobe of the left lung, the inferior lobe of the left lung, the superior lobe of the right lung, the middle lobe of the right lung and the inferior lobe of the right lung. The lungs may be divided into more than five regions. Each lung region may be defined by at least one voxel in a 3-dimensional image of the lung. Each region may be defined by a plurality of voxels in a 3-dimensional image of the lung.

[0031] The step of calculating a relative regional distribution of air in the lung may be performed based on at least one of: calculating a relative regional shape of the lung; calculating a relative regional density of the lung; calculating a relative regional size of the lung; calculating a relative regional expansion of the lung.

[0032] Each of the 3-dimensional images of the lungs may be acquired using computerized tomography (CT) technique. Each of the 3-dimensional images of the lungs may be acquired using Magnetic Resonance Imaging (MRI). Each of the 3-dimensional images may be acquired in the inspiration breathing phase. At least one of the images may be acquired at a peak inspiration breath hold.

[0033] The images may be acquired during different imaging sessions. [0034] A visual image may be created from the 3-dimensional image datasets.

[0035] The step of comparing the relative regional distributions of air in the lung calculated at each breath to assess a change in relative regional distribution of air in the lung between the two breaths may comprise at least one of the steps of: normalizing the two 3-dimensional images; modifying at least one of the 3-dimensional images; dividing the images into at least two regions; deforming at least one of the images.

[0036] The method being performed without requiring a visual image, whether digital, physical or otherwise, to be produced.

[0037] Further embodiments provide a method of identifying changes in lung health using 3- dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquiring two 3-dimensional image datasets of a lung, each image dataset being acquired during a different breath of a lung; comparing the image datasets to assess a change in relative regional distribution of air in the lungs between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

[0038] In a further embodiment a system for identifying changes in lung health using 3- dimensional image datasets, including: an image acquisition module, configured to acquire two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath of a lung; a processor configured to compare the image datasets to assess a change in relative regional distribution of air in the lung between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

[0039] In a further embodiment a non-transitory computer readable storage medium having a computer program stored therein, that when executed by a processor of a computer, causes the computer to execute steps directed to identifying changes in lung health using 3-dimensional image datasets, including: acquiring two 3-dimensional image datasets of a lung, each image dataset being acquired during a different breath of a lung; comparing the image datasets to assess a change in relative regional distribution of air in the lungs between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health. [0040] In a further embodiment a method of identifying changes in lung health using 3- dimensional images, including acquiring two 3-dimensional image datasets of a lung, each 3- dimensional image dataset being acquired during a different breath of a lung; normalizing the two three-dimensional image datasets to produce two normalized 3-dimensional image datasets, comparing the two normalized 3-dimensional image datasets to assess a change in relative regional distribution of air in the lung between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

[0041] A further embodiment provides a method of identifying changes in lung health using 3- dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath of a lung; normalizing the two three-dimensional image datasets to produce two normalized 3-dimensional image datasets, comparing the two normalized 3-dimensional image datasets to assess a change in relative density of the lung between the two breaths, wherein the change in relative regional density of the lung is indicative of a change in lung health.

Brief Description of the Figures

[0042] In order that the invention be more clearly understood and put into practical effect, reference will now be made to preferred embodiments of an assembly in accordance with the present invention. The ensuing description is given by way of non-limitative example only and is with reference to the accompanying drawings, wherein:

[0043] Figure 1 is a schematic diagram showing the basic design of a CT system.

[0044] Figure 2 is a schematic diagram of a CT system.

[0045] Figure 3 is an example of a CT image.

[0046] Figure 4A is a flow diagram showing the process of using CT images to identify a change in lung health.

[0047] Figure 4B is a flow diagram showing the process of using CT images to identify a change in lung health.

[0048] Figure 5 shows examples of a first CT image of a lung and a second CT image of a lung. [0049] Figure 6 shows the volume of various regions of the lungs taken from a first CT scan on a patient.

[0050] Figure 7 shows the volume of various regions of the lungs of Figure 6 as a percentage of the total volume of the lungs.

[0051] Figure 8 shows the volume in various regions of the lungs taken from a second CT scan on a patient.

[0052] Figure 9 shows the volume of various regions of the lungs of Figure 8 as a percentage of the total volume of the lungs.

[0053] Figure 10 shows the comparison between the percentage data of Figures 7 and 9.

[0054] Figure 11 shows the percentage changes for each region of the lung.

[0055] Figure 12 maps the equivalent regions of the lungs shown in Figures 6 and 8.

[0056] Figure 13 is a graphical representation of the change in volume in the lobes of the lungs between Figures 6 and 8.

[0057] Figure 14 shows the transformation of Image 1 (2000 ml) to Image 1 ’ (3000 ml).

[0058] Figure 15 shows a comparison of the volumes of the various lobes in Image 1’ with Image 2.

[0059] Figure 16 shows any change in volume between the lobes shown in Figure 15.

[0060] Figure 17 shows the changes in volume in each of the lobes represented as a percentage of the overall volume of the lungs.

[0061] Figure 18 shows the number of voxels in each of the five regions in a first CT image.

[0062] Figure 19 shows the number of voxels in each of the five regions in a second CT image. [0063] Figure 20 shows a comparison of the percentage number of voxels in each region of the lung between a first and second CT image.

[0064] Figure 21 is a schematic block diagram of an apparatus for identifying changes in lung health.

[0065] Figure 22 shows the volume of a region of the lungs taken from a first CT scan on a patient.

[0066] Figure 23 shows the volume of a region of the lungs of Figure 22 as a percentage of the total air in the lungs.

[0067] Figure 24 shows the volume of a region of the lungs taken from a second CT scan on a patient.

[0068] Figure 25 shows the volume of a region of the lungs of Figure 24 as a percentage of the total volume of the lungs.

[0069] Figure 26 shows the comparison between the percentage data of Figures 23 and 25.

[0070] Figure 27 shows the percentage change for the region of the lung.

Detailed Description:

[0071] Embodiments use 3-dimensional (3D) images of the lungs to assess a change in lung health. Detailed 3-dimensional (3D) images of internal body tissue may be obtained using X-ray imaging apparatus, including 3D CT (computed tomography) scanners. These non-invasive techniques provide in vivo internal detailed images of human organs. CT imaging offers good resolution and penetration for medical imaging.

[0072] CT is a well-known technique used to reconstruct an object in three-dimensional space from two dimensional projections. Computerized tomography (CT) scans capture multiple X-ray images at different angles around a body. This produces X-ray images of the body from many different angles. The system processes the data in the images and combines the images to produce a three-dimensional representation of the body and can create cross-sectional images through the body. These slices show bones, blood vessels, soft tissues and other internal structures inside the body. [0073] Referring now to Figure 1 , a schematic diagram outlining the basic design of a CT system is shown. The CT system includes a circular CT scanner 20 including a rotating X-ray tube assembly and a detector array positioned diametrically opposite the X-ray tube assembly on the other side of the circle. The CT system includes an imaging bed 30 for a patient to lie on during the scanning procedure. Imaging beds may include securing wraps 40 or straps to help hold the patient still during scanning. The imaging bed 30 is typically movable along the axis of the CT scanner. During scanning, a patient lies on the imaging bed 30 and is positioned within the CT scanner 20. When the patient is in position, the X-ray tube assembly and detector array rotate around the patient and capture a number of X-ray images of the patient at different angles. The X-ray images are processed using computer algorithms to produce a three- dimensional reproduction of the inside of the patient. Two-dimensional slices through the three- dimensional reproduction of the inside of the patient can be created.

[0074] Typically, integrated object density in the projection direction is calculated from the X-ray attenuation, which will be proportional to the pixel intensity values on a digital projection image. The object structure is reconstructed from projection images taken at different viewing angles, using Fourier back-projection or algebraic methods. Variants also exist for reconstruction of objects from few projection angles, which use iterative methods to reconstruct the sample's structure, often exploiting prior knowledge of the sample, for example that it is made up of a single material.

[0075] The basic components of a CT scanner are shown in Figure 2. Figure 2 shows a cross- sectional view through a CT scanner 200 with a patient 270 positioned within the scanner 200. An X-ray tube assembly includes X-ray tube 210, X-ray filter 220 and X-ray collimator 230. X- ray tube 210 is positioned within CT scanner 200 and rotates around the central axis of the CT scanner 200. X-ray filter 220 is positioned in front of the x-ray tube 210 and is tuned to absorb particular energy x-rays. Generally, X-ray filters in CT scanners are tuned to absorb low energy X-rays which have high attenuation and can cause blurring in images and reduce contrast detail. X-ray tube collimator 230 aligns and directs the X-rays towards the area of interest of the patient. The X-ray beam is represented as 240 in Figure 2. The collimator 230 is typically configured so that the brightest region of the X-ray source is centred on the sample region of interest.

[0076] A detector array 250 is positioned diametrically opposite the X-ray tube assembly. The X-ray detector 250 detects X-ray radiation which has been attenuated by tissues of the patient’s body and converts it into a digital signal. Modern X-ray detector arrays have multiple rows of detectors, for example between 8-64 rows. Each row includes many detectors, for example between 1000-2000 detectors. Some important characteristics of X-ray detectors include having high detection efficiency, being small and efficiently arranged to limit the distance between detectors in the array to increase the overall resolution of the image and having a fast response time.

[0077] The X-ray tube assembly and the detector array are positioned within a cylindrical gantry 260. The imaging bed typically moves along the central axis of the gantry. The patient lies on the imaging bed and the clinician controls the position of the bed within the gantry to bring the relevant part of the patient’s body into the imaging area of the X-ray tube. The X-ray tube and the detector array spin within the gantry around the patient to capture X-ray images at different angles.

[0078] The images captured by the detector array are computer processed and combined to produce a three-dimensional representation of the inside of the patient.

[0079] The CT scanner is controlled by a central control system which is managed via a user interface. The user interface typically allows alignment, imaging and analysis. The central control system and user interface preferably allows simple activation of technical functions such as testing, calibration and alignment. The interface preferably also allows control of other user related functions such as imaging, image processing and visualisation of reconstructed results.

[0080] CT scans are commonly used for imaging lungs. CT scans may be taken while the patient is continuously breathing or while the patient is holding their breath (during a breath hold). Preferably, inspiratory breath-hold CTs taken at peak inspiration are used to capture a CT image of the lungs. The patient takes an inspiratory breath to fill the lungs with air, holds the breath and remains still for the duration of the scan. The scan may last a few seconds as the X- ray assembly and detector rotate around the patient. This allows the chest to be scanned from multiple angles while in the same inspiratory state and in the same position. Since the lungs remain in the same air-filled state during the scan and for each of the images taken from different angles, the images can be combined and blurring is reduced.

[0081] In typical CT reconstruction techniques, integrated object density in the projection direction is calculated from the X-ray transmission, which will be proportional to pixel intensity values on a digital projection image. In the case of a material of constant density, this integrated object density will be proportional to the object thickness. The contrast of the particle speckle (defined as the ratio of the standard deviation of the Image Intensity to the mean intensity) will increase with the square root of object thickness and so this statistic may also be used for tomographic reconstruction of the object’s structure. This is advantageous, as in many cases, including in vivo imaging of blood vessels the absorptions contrast alone is insufficient for tomographic reconstruction.

[0082] The output from a CT scan is a CT dataset. The dataset is a dataset representing a 3- dimensional image of the CT scan. The dataset may be referred to as a 3-dimensional image dataset. The 3-dimensional image dataset may be represented visually. This visual representation is typically used by clinicians to help observe lung or other body part. The 3- dimensional dataset may be represented in a digital form, for example on a display screen of a digital device, for example a computer monitor, tablet screen, mobile phone screen etc. The 3- dimensional image dataset may be reproduced into a physical form. In the description, the terms CT image, image, 3-dimensional image, CT image dataset, or similar may be used to refer to the 3-dimension image dataset.

[0083] The embodiments described identify changes in lung health of a patient using data obtained from CT imaging. In further embodiments data may be obtained from other imaging techniques, for example Magnetic Resonance Imaging (MRI).

[0084] Figure 3 shows an example CT image. The image shows a cross-sectional slice through the chest of a patient between the neck and the waist. A CT produces an image of the full 3- dimensional volume, the image of Figure 2 is a single slice through that 3-dimensional volume.

[0085] CT images are presented in black and white with varying shades of greyscale in between. These shades between black and white are referred to as the image intensity, but may also be referred to as shade, colour or brightness. The image intensity of a particular tissue feature in the image is determined by the degree to which it absorbs (attenuates) x-rays. Features of higher density have higher attenuation coefficients. The image intensity of a feature displayed in the image is determined by its mean attenuation. CT images use Hounsfield Units, and tissues with a high Hounsfield score have a high attenuation coefficient and appear white on the CT image (for example bones). Those features with a low Hounsfield score have a low attenuation coefficient and appear dark on the CT image (for example low density tissue, such as aerated lungs). CT scanners are calibrated so a particular density is always represented at the same image intensity. This allows calculations of density for the body parts within the image and comparison between images.

[0086] In the example image of Figure 3, the bone of the ribs is shown at 310 and appears white in the image due to its high attenuation coefficient. The body is shown at 320 and appears light grey, having a lower attenuation coefficient. Blood vessels 330 appear as light grey. The lung 340 appears in various degrees of dark grey having a low density. Airway 350 is the darkest feature of the image having the lowest density due to the high air content.

[0087] Embodiments of the system identify changes in lung health of a patient using CT imaging. Two 3-dimensional CT image datasets of a lung are acquired where each image dataset is acquired during a different breath. This may be referred to as one image dataset being acquired in a subsequent breath to the other image dataset. The two image datasets are acquired during imaging sessions on the patient. For example, the image datasets may be acquired in two different imaging sessions that are hours apart, days apart, weeks apart, months apart or years apart. The process may also be used to assess changes in lung health by comparing image datasets acquired in a single imaging session, where the CT images may only be minutes apart. Embodiments compare the image datasets to assess a change in the relative regional volume of one or more regions of the lung between the two breaths.

[0088] Regional volume distribution measures the distribution of volume of different regions within a lung. Regional volume distribution can be calculated over small regions, for example on a voxel-by-voxel basis on the image, or averaged over a larger regions of the lung including multiple voxels. Regions of the lung may be defined based on anatomical structure of the lung, such as the left and right lungs, lobes of the lungs, or the sub-lobes of the lungs (sometimes referred to as segments). Alternatively, regions may be defined based on individual voxels, groups of voxels defined within the images. Regions of the lung may be defined geometrically, such as cubes, spheres, cones or other conical structures, etc. The physical volume of the regions may be less than 1500mL, less than 1000mL, less than 500L, less than 250mL, less than 100mL, less than 50mL, less than 30mL, less than 20mL, or less than 10mL. Other arbitrary techniques for defining the regions of the lungs may be used.

[0089] Relative regional volume distribution defines the volume of a lung region with respect to the total volume of the lung. The volume of a region in the lung is indicative of the amount of air in that region of the lung. Relative regional volume distribution may define the volume in different regions of the lung as a proportion or fraction or percentage of the total volume of the lung. This allows the distribution of air throughout the lung to be calculated relatively with respect to the total volume of the lung. Since, for each scan, the volume of the region of the lung is calculated relatively with respect to the total volume of the lung, the distribution of volume of the lungs can be compared and assessed for CT images captured during different breaths. Although the total volume of the lung may be different between the two CT images, the relative distribution of volume of the lung can be compared. The relative distribution of volume of the lung can provide information about the performance of the lung. Changes in relative distribution of volume of the lung may indicate a change in lung health. In particular, knowing how the distribution of volume of the lung has changed over time (i.e. between the scans) can provide important information about change in lung function and lung health. In other words, determining the volume of a given region in two CT scans, calculating the relative volume of the region in each of the CT scans, and then comparing the relative volume to quantify the change in relative volume, allows a clinician to determine whether there has been a redistribution of air in the lungs, and thus a change in the lung function of the patient.

[0090] It is well understood that the volume of the lung, or any lung region, in a CT image includes both the volume of the lung tissue itself, and the volume of the air within that lung tissue. Calculation of the volume of the lung, volume of a region, relative volume of a region, etc provides an indication of the volume of air present in the lung, region, etc. Also, any change in relative volume of a region provides an indication of relative volume of air in a region. The distribution of volume of a lung is an indication of the distribution of air within the lung.

[0091] When the lung is filled with air, for example during inhalation and in particular as a lung approaches peak inhalation, the volume of air is much greater than the volume of tissue. Accurate measurements of air in the lungs require the volume of tissue to be accounted for, but it is a reasonable approximation that the volume of lung (or volume of a region of the lung) is the volume of air (or volume of air in the region). Consequently, terms relating to volume of lung and volume of air may be used interchangeably within this specification for the purposes of describing the invention, particularly in relation to data retrieved from images captured during an inspiration breath hold by a patient. Similarly, terms relating to distribution of volume of the lung or distribution of air in the lung may also be used interchangeably.

[0092] By calculating the relative regional volume distribution of the lungs, meaningful comparisons can be made between images captured on patient lungs containing different volumes of air, i.e. having different overall volumes, because a comparison is made based on the relative regional volume distribution of air rather than absolute regional volume distribution of air. By comparing relative volume, differences due to differences in the absolute volume of air in the lungs in each image are normalized and so a meaningful comparison of lungs holding different volumes can be made. This allows differences in regional distribution of the volume of air to be identified. The difference in relative regional volume distribution of air may be identified by assessing relative changes in several factors determined from the images, including changes in the volume of air, changes in the density of lungs, changes in the shape of the lungs, changes in the expansion of the lungs.

[0093] By comparing relative regional volume distribution of the lung, areas of the lungs that exhibit different performance of different function in the different images can be identified. This change in performance or function may be indicative of a change in lung health.

[0094] The relative volume of a region of the lung may also be referred to as: the proportional volume of the region; or, the percentage volume of the region; or, the fractional volume of the region; when compared with the volume of a larger part of the lung, for example the entire lungs. This step of calculating the volume of the region as a relative volume, proportional volume, percentage volume, fractional volume, etc., has the effect of normalising the volume data for the region and allows the volume of the region to be compared for different breaths in which the volume values (i.e. the volume of the region in litres) for different breaths may be different.

[0095] The process of using CT images to identify a change in lung health is now described with respect to the example processes of Figures 4A and 4B. The process described is typically performed in software and executed on a computer system. An example system for implementing the method for identifying changes in lung health is described with respect to Figure 21. Although the process described uses data acquired from CT images, in other embodiments lung image data may be acquired using other imaging techniques, for example MRI.

[0096] Figure 21 is a schematic block diagram of an apparatus 2100 for identifying changes in lung health. As described above, images of the patient are captured by CT scanner 2110. The data from the CT scan may be displayed visually, either in 3-dimensions or as a 2-dimensional slices (typically axial slices) through the 3-dimensional image or retained as a data file.

[0097] Apparatus 2100 is configured to perform the methods described to identify changes in lung health using the CT scan data. The components of apparatus 2100 may be co-located or may be distributed and form part of a distributed computing system. In the case of a distributed computing system, components may be connected across communication networks, for example a mobile communication network. The distributed computing system may be referred to as a cloud computing system. Apparatus 2100 is configured to perform data processing on the CT scan data, the data processing may include the processes of data deformation, data normalization, data comparison as well as other image processing or data processing steps.

[0098] The apparatus 2100 may include one or more network interfaces 2120 that may facilitate communication between the apparatus 2100 and one or more other apparatuses using any suitable communications standard. For example, the interface 2120 may enable the receipt of image datasets from imaging apparatus 210, where the image datasets represent images captured by the imaging apparatus. In the example of Figure 21 , imaging apparatus 210 is a CT scanner. The interface 2120 may also enable the receipt of further information relating to the patient and/or the image datasets. The interface 2120 may be a LAN interface that implements protocols and/or algorithms that comply with various communication standards of the Institute of Electrical and Electronics Engineers (IEEE), such as IEEE 802.11 , while a cellular network interface implement protocols and/or algorithms that comply with various communication standards of the Third Generation Partnership Project (3GPP) and 3GPP2, such as 3G and 4G (Long Term Evolution), and of the Next Generation Mobile Networks (NGMN) Alliance, such as 5G.

[0099] The apparatus 2100 may include one or more processors 2130 configured to access and execute computer-executable instructions stored in at least one memory 2140. The processor 2130 may be implemented as appropriate in hardware, software, firmware, or combinations thereof.

[0100] Processor 2130, implemented in hardware may be a general-purpose processor. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor 2130 may include, without limitation, a central processing unit (CPU), a digital signal processor (DSP), a reduced instruction set computer (RISC) processor, a complex instruction set computer (CISC) processor, a microprocessor, a microcontroller, a field programmable gate array (FPGA), a System-on-a-Chip (SOC), or other programmable logic, discrete gate or transistor logic, discrete hardware components, or any combination thereof, or any other suitable component designed to perform the functions described herein. Processor 2130 may also include one or more application- specific integrated circuits (ASICs) or application -specific standard products (ASSPs) for handling specific data processing functions or tasks. Processor 2130 may also be implemented as a combination of computing components, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP, or any other such configuration.

[0101] Software or firmware implementations of processor 2130 may include computerexecutable or machine-executable instructions written in any suitable programming language to perform the various functions described herein. Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise. The software may reside on a computer-readable medium. A computer-readable medium may include, by way of example, a smart card, a flash memory device (e.g., card, stick, key drive), random access memory (RAM), read only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), a general register, or any other suitable non-transitory medium for storing software.

[0102] The memory 2140 may include, but is not limited to, random access memory (RAM), flash RAM, magnetic media storage, optical media storage, and so forth. The memory 2140 may include volatile memory configured to store information when supplied with power and/or non-volatile memory configured to store information even when not supplied with power. The memory 2140 may store various program modules, application programs, and so forth that may include computer-executable instructions that upon execution by the processor 2130 may cause various operations to be performed. The memory 2140 may further store a variety of data manipulated and/or generated during execution of computer-executable instructions by the processor 2130.

[0103] Memory 2140 may store various program modules, application programs, and so forth that may include computer-executable instructions that upon execution by the processor 2130 may cause various operations to be performed. For example, the memory 2140 may include an operating system module (O/S) that may be configured to manage hardware resources such as network interface 2120 and provide various services to applications executing on the apparatus 2140. [0104] Memory 2140 includes storage modules. Application storage module 2141 stores applications for identifying changes in lung health. Memory 2140 stores additional program modules which may be called and executed during execution of the application. Additional storage modules may include data storage module 2142 for storing CT image datasets for comparison. Mapping module 2143 may include algorithms and programs for conducting image processing for example including transformation or deformation algorithms including deformable image registration (DIR) algorithms; or, particle image velocimetry (PIV), in order to map voxels from different images corresponding to equivalent parts of the lung.

[0105] Normalisation module 2144 includes software programs for normalizing the CT datasets, for example by comparing a region of lung tissue to the total lung tissue in order to normalize any calculated parameters for that region.

[0106] Comparison module 2145 includes software programs for comparing datasets and identifying differences between those datasets, for example to compare a region of one CT scan to a corresponding region of another CT scan.

[0107] Each of the modules includes functions in the form of logic and rules that respectively support and enable the various functions described herein with reference to the Figures, including deforming the images and image datasets, normalizing the images and image data sets, comparing the image data sets and/or images, dividing the image data sets and/or images. Although illustrated as separate modules in Figure 21 , one or more of the modules may be a part of or a submodule of another module. For example, the mapping module 2143 may be a submodule of normalization module 2144.

[0108] The modules 2141 , 2142, 2143, 2144, 2145 disclosed herein may be implemented in hardware, or software and/or firmware implementations executed on a hardware platform. The hardware may be the same as described above with respect to the processor 2130. Likewise, the software and/or firmware implementations may be the same as described above with respect to the processor 2130.

[0109] User interface 2110 facilitates user interaction with apparatus 2100. User interface includes a user input module to receive input from the user. The user input module may include a keyboard, touchscreen, touch pad, microphone or other input device. User interface allows a user to select applications for execution, datasets for comparison and other modules for execution by the processor. User interface 2110 may also include an output module. The output module may include a display or other means for conveying information to the user.

[0110] The process of using CT images to identify a change in lung health is now described with respect to the example processes of Figures 4A and 4B, executed on the example system of Figure 21.

[0111] Referring to Figure 4A, at 410 two CT images of the lungs of a patient are acquired from a CT scanner 210 for comparison. These two CT images are received at apparatus 2100 via network interface 2120. The images may be selected manually by a user via user interface 2110. The CT images are defined by digital data representing x-ray attenuation. Reference to CT images may relate to CT image data visibly produced, for example in digital form for display on a screen (as shown in Figure 5) or in a hard copy or may relate to the CT data itself.

[0112] The images may have been acquired at different times in different scanning sessions. For example, the images may have been acquired hours, days, weeks or months apart. The images may have been captured using different CT machines, for example in a different clinic or hospital. In some cases, the patient may have undergone treatment between the imaging sessions. The patient may have reported a change in respiratory symptoms.

[0113] Preferably the images are captured in the same breathing phase, for example both images may be recorded in the inspiratory phase or in the expiratory phase. Preferably the images are recorded during an inspiratory breathing phase. Preferably the images are recorded during a breath hold. Preferably the images are recorded during a maximum inspiratory breath hold.

[0114] The total volume of air in the lungs may be different in both CT images. For example, in the first image the patient may have inhaled more deeply or with more vigor or effort resulting in inhaling a higher volume of air into the lungs.

[0115] Preferably the images are acquired such that the axial slices represent the same slice through the lungs (e.g. the patient is located at the same position, angle, or both so that the position and angle through the lungs is the same for both CT images).

[0116] Figure 5 shows examples of a first CT image of a lung and a second CT image of a lung. The CT images are taken on the same patient. In the example of Figure 5 Image 2 was recorded 6 months after Image 1 . Image 1 and Image 2 are CT scans of a patient’s lungs. Both images are recorded during an inspiratory breath hold (also referred to as inspiration breath hold). Image 1 and Image 2 are typical of images recorded on patients with respiratory conditions in respiratory clinics or hospitals.

[0117] In some examples, there is an assumption that the lung fills with air homogeneously. This assumption provides that for a given CT image dataset the regional distribution of air in the lung is the same regardless of the volume of air within the lung, so the regional distribution of air in the lung is the same at 60 % capacity and 90% capacity. This assumption may be more reliable for images captured at more similar capacities, for example within 10%.

[0118] After selection of the two CT images for comparison, at step 420 the images are analysed. The CT images are analysed using application software stored within application storage module 2141 and executed using processor 2130. The application software may be selected and initiated by a user via user interface 2110. As described above, the CT images may be compared by manipulating and/or comparing the data defining the CT images.

[0119] In the example of Figure 5, the lungs have been divided into five lobes according to the anatomical structure of the lungs. The lobes of the lung are well defined in lung anatomy. The left lung 5110 5210 includes two lobes, namely the left superior lobe 5111 5211 and the left inferior lobe 5112 5212. The right lung 5120 5220 has three lobes, namely the right superior lobe 5123 5223, right middle lobe 5124 5224, and right inferior lobe 5125 5225.

[0120] Physicians can make qualitative visual assessments in several ways. For example, physicians can use the intensities of the lobes within the same image to determine the relative distribution of air within the lungs. For example, in Image 1 , the superior lobe of the left lung 5111 and the inferior lobe of the left lung 5112 have different intensities. These different intensities indicate that the lobes have different densities. In Image 1 , the inferior lobe of the left lung 5112 is darker than the superior lobe of the left lung 5111. The different intensities indicate that the density of the superior lobe 5111 is greater than the density of the inferior lobe 5112. This may indicate that more air is present in the inferior lobe 5112, having a darker intensity, compared with the superior lobe 5111. This comparison provides valuable information about the distribution of air within the lungs in a single CT scan.

[0121] When the physician performs a similar qualitative visual assessment of the left lung in Image 2, a different distribution of air is observed. In Image 2, the intensity of the superior lobe 5211 and the inferior lobe 5212 are more similar compared with the lobes of the left lung in Image 1 . This similarity of intensity indicates that in Image 2, for the left lung, the density of the superior lobe 5211 is similar to the density of the inferior lobe 5212. This similarity of density indicates that air may be distributed quite evenly between the superior lobe and the inferior lobe in Image 2. This distribution of air within the left lung in Image 2 is different from the distribution of air in the left lung in Image 1 , which indicates that there is relatively more air present in the inferior lobe 5112. In Image 2 there is a regular distribution of air throughout the left lung, indicated by the consistent intensity in Image 2. In Image 1 there is a difference in distribution of air, with the superior lobe appearing to contain less air than the inferior lobe. This difference in distribution of air within the left lung may be an indication of a change in lung health of the left lung between the images.

[0122] A similar assessment may be made on the right lungs and also across the left lung and the right lung. This assessment of distribution of air within the lungs can provide information about the operation and health of the lungs. When CT scans of a patient are compared, for example CT scans taken at different times, or before and after treatment, changing operation and health of lungs can be monitored.

[0123] While an idealized example a visual assessment of the distribution of air in the lungs can be performed (as the total lung volume remains the same), in real life examples the volume of the two images can be quite different. This difference in volume can lead to situations in which there is more air in a given lobe in one of the CT scans (implying that in that scan a patient has improved that lobe’s lung function), however, due to the overall increase in volume of the CT scan there is actually a relative reduction in the air in that lobe. By calculating and monitoring the change in relative regional distribution of air within the lung, better information can be obtained about lung health and any changes in lung health between the CT scans.

[0124] For comparison the data from different images is processed so it can be directly compared. In one example, the system calculates the proportion of the total volume of the lungs in each of the regions and compares the proportions of each region. The proportion may be calculated as a percentage or as a fraction of the total volume of the lung, for example by dividing the volume of the region by the total volume of the lung. This proportional volume of the region compared with the total volume of the lung may be referred to as the relative volume of the region of the lung. Normalizing the data makes it directly comparable with values from other data sets, and allows for the quantification of the change in relative volume. Alternatively, normalization can be performed globally by scaling the total volume of the lungs in one CT scan to the total volume of the lungs in the other CT scan. For example, if during a first CT scan the lungs held 2000 ml and in a second CT scan the lung held 3000 ml of air, the data from the CT scans could be adjusted to make each scan include the same volume of air, in order to allow a direct comparison.

[0125] At 430 differences in the images are identified to assess a change in lung health of the patient between the images. Differences are identified by determining a change in the relative regional volume distribution of air in the lungs between the images. A change in relative regional volume distribution of air in the lung may be indicative of a change in lung health.

[0126] An example of the process for comparing the CT images is now described with reference to Figure 4B.

[0127] Due to the deformation of the lungs during respiration, CT scans taken at different volumes may not be able to be directly compared as the anatomy of the lung might be different and in a different location in the two images. In order to allow for a proper comparison of the same region of lung tissue, the region of lung tissue of interest must be mapped between the two CTs. One method for achieving this is to deform one CT to the other CT, and at 4210 the images are deformed to transform the data sets into a common coordinate system to allow proper comparison of different CT datasets. Deformable image registration (DIR) is a process commonly used in medical imaging to transform different data sets into one common coordinate system. The DIR process can account for changes in the shape of anatomy between the images. The common coordinate system may be the coordinate system of one of the images or a separate coordinate system. After deformation, the respective physical positions in the two images overlap/correspond correctly. This process defines the relationship of voxels in one CT to the voxels in the other CT, thereby allowing for the same region of lung tissue to be interrogated in each of the CTs.

[0128] The deformation algorithms are stored in mapping module 2143. The deformation algorithms are called by the application storage executed by processor 2130.

[0129] Deformable image registration algorithms can be applied to deform one image to another. The regions of the images can be mapped together. In the case of the lungs, DIR techniques allow the same voxels to be mapped together between two images so for a particular voxel its position and size can be identified in each image. By deforming the anatomy of one image to the anatomy of the other, variation of the angle or configuration of the lungs between the images can be compensated for and the images can be compared. The mapping process allows for the volume of the region of tissue to be calculated in both of the CT images.

[0130] Other mapping techniques may be used to determine the relationship of voxels in one CT to the voxels in the other CT, including particle image velocimetry (PIV).

[0131] Figure 6 shows the volume in various regions of the lungs taken from a first CT scan on a patient. As described above, this is an approximation of the volume of air in various regions of the lung. The images are divided up into regions to define various regions of the lungs at step 4220 and the equivalent regions of each image may be compared. Different techniques may be used to divide up the lungs. The images may be divided based on anatomical structure of the lung, for example divided into lobes or sub-lobes. The images may be divided into regions based on individual voxels, groups of voxels. The groups of voxels may be in groups of any suitable size, and can vary depending on the resolution of the CT scan. The number of voxels in a region can be at least 2 voxels, at least 4 voxels, at least 5 voxels, at least 8 voxels, at least 27 voxels, at least 64 voxels, at least 125 voxels, at least 216 voxels, at least 500 voxels, at least 1 ,000 voxels, at least 10,000 voxels, at least 100,000 voxels, at least 200,000 voxels, or at least 1 ,000,000 voxels. The number of voxels in a region may also be no more than 2 voxels, no more than 4 voxels, no more than 5 voxels, no more than 8 voxels, no more than 27 voxels, no more than 64 voxels, no more than 125 voxels, no more than 216 voxels, no more than 500 voxels, no more than 1 ,000 voxels, no more than 10,000 voxels, no more than 100,000 voxels, no more than 200,000 voxels, no more than 1 ,000,000 voxels. Alternatively, the images may be divided geometrically, such as cubes, spheres, cones or other conical structures, and could be chosen to lie on a regular grid. Other arbitrary techniques for dividing the images may be used. Examples of other techniques include geometric-type techniques dividing the images by blocks or curves.

[0132] In the example of Figure 6 the images are divided into five lobes according to the anatomical structure of the lungs. This division of the lungs uses the same regions as described with reference to Figure 5 above. The lobes of the lung are well known and well defined in lung anatomy. The left lung includes two lobes, namely the superior lobe and the inferior lobe. The right lung has three lobes, namely the superior lobe, middle lobe, and inferior lobe.

[0133] In other examples, the lung may be divided into a different number of regions. For example, on a voxel-by-voxel basis. For example, the lung could be divided into at least 5 regions, at least 6 regions, at least 10 regions, at least 15 regions, at least 18 regions, at least 19 regions, at least 20 regions, at least 25 regions, at least 40 regions, at least 50 regions, or at least 100 regions.

[0134] After the images have been mapped (via the DIR process), each of the regions in each of the images may be normalized at 4230. By normalizing the regions, the CT images can be compared directly. For example, by normalizing the left superior lobe by the total volume of the lungs (i.e. the sum of the volume of the 5 lobes) it is possible to determine the proportion of the overall lung volume that the left superior lobe occupies, which can be considered as either a percentage or a unitless number. By also normalizing the left superior lobe in the second CT it is possible to directly compare how the relative volume of the lobe has changed between the two scans, as the impact of the different total lung volumes for the scans has been accounted for and removed. The images may be normalized using several different mathematical techniques, and the invention is not limited to any particular method of normalizing the images. Some of the following methods for normalization are based on total volume of air in the lungs. Other methods of normalizing the images are based on other factors. It is also possible to normalize the volume of the region to a differently defined volume to the entire lungs, for example the region could be normalized to its respective lung (e.g. the left lung), the respective lobe (e.g. for the anterior upper left sub-lobe this would be the left superior lobe), or any other suitable volume for normalizing the volume data. Typically, the volume of the region is normalised to the volume of a defined larger part of the lung, for example the entire lungs.

[0135] In a first series of examples the CT image data is normalized by calculating the total volume of the lungs for a CT image and normalizing each of the regions in the CT image dataset based on total volume of the lungs during each CT scan. In other examples the volume of only selected regions of the lung is normalised to the total volume of the lungs. In such examples not all regions of the lung are required to be analysed and normalised. Instead, the volume of selected regions is determined and normalised to the total volume of the lung. Any change in the relative regional volume of a selected region is an indication that the distribution of air in the lungs has changed and may indicate a change in lung health.

[0136] Alternatively, rather than calculating the volume of the lobe and normalizing by the volume of the lung, it is possible to use the Hounsfield Unit information in the CT to create the relative measurements. For example, the average Hounsfield Unit value of the lobe could be calculated, and this could then be normalized by the sum of the average Hounsfield Units for the five lobes. [0137] At 4240 the relative volume of each region is calculated. At 4250 the relative volume of corresponding regions in different images are compared.

[0138] Software programs for dividing the CT images into regions, and normalizing the regions in the CT images, may be stored within normalization module 2144. These software programs may map voxels from different images corresponding to equivalent parts of the lung and normalize the data based on an equivalent volume.

[0139] The CT images are compared using software programs for comparing datasets and identifying differences between those datasets which may be stored in comparison module 2145. By comparing the normalized volume difference of the region it is possible to quantify any relative change in the volume of the region between the two CT scans.

[0140] In the following examples, and with reference back to Figure 5, the volume of the lungs in Image 2 is 3000 ml (see Figures 8 and 9) and the volume of the lungs of Image 1 is 2000 ml (see Figures 6 and 7), i.e. the patient’s lungs contained an extra 1000ml of air when scanned for Image 2 compared with when scanned for Image 1 .

[0141] A first example of calculating the change in regional volume distribution of the lungs between two CT scans based on calculations of total volume of the lungs is described with reference to Figures 6 to 11 . In this first example in the images the lungs are divided into 5 lobes, the superior and inferior lobes of the left lung and the superior, middle and inferior lobes of the right lung. It is well understood that the volume of the lung, or any lung region, in a CT image includes both the volume of the lung tissue itself, and the volume of the air within that lung tissue.

[0142] Figure 6 is a schematic figure representing the volume of the lobes of the lungs calculated from a first CT scan on a patient. The volume of each of the lobes is calculated from the CT scan using known techniques. The number of voxels in each lobe of the lung, as measured in the CT image data, is used to calculate volume. This is possible as the physical volume of the voxels in the CT is known, and therefore the absolute volume can be calculated (i.e. the volume in mL or L, rather than the volume in voxels). The lung has been divided into 5 lobes, as described above. The volume of each lobe is shown in Table 1 . In the left lung, the left superior lobe includes 200 ml and the left inferior lobe includes 700 ml. In the right lung the right superior lobe contains 300 ml , the right middle lobe contains 360 ml and the right inferior lobe includes 440 ml . The total volume of the lungs of the patient in the first CT scan is 2000 ml. The total volume of the lungs can be calculated by either summing the volume for all of the regions of the lung, or it can be independently calculated, for example by summing all of the voxels in the lung.

[0143] In the first example the percentage of the total volume of the lungs in each of the lobes is calculated, for comparison. For each lobe, the percentage of each lobe is calculated as a percentage of the total volume of the lungs, 2000 ml. This is calculated for each lobe by dividing the volume of the lobe by the total lung volume, and expressing the result a percentage. For example the volume of the left lung the superior lobe is 200 ml. 200 ml represents 10% of the total volume of the lung of 2000 ml. Figure 7 shows the volume of each lobe converted to a percentage of the total the lung volume. By converting the volumes to percentages the data can easily be compared with other CT scans. The percentage of the total volume of each lobe is presented in Table 1. As described above, these volumes during an inspiration breath hold are approximations of the volume of air.

[0144] In the left lung, the left superior lobe includes 10 % of the total volume of the lung and the left inferior lobe includes 35 % of the total volume of the lung. In the right lung the right superior lobe contains 15 % of the total volume, the right middle lobe contains 18 % of the total volume and the right inferior lobe includes 22 % of the total volume. These percentages can now be directly compared with the percentages of each lobe in other CT scans of the patient to assess whether there is a relative change in the regional volume distribution of the lungs of the patient (i.e. to determine if there has been a redistribution of air/ventilation in the lung between the two scans).

Table 1.

[0145] Figure 8 is a schematic figure representing the volume of the lobes of the lungs calculated from a second CT scan on the patient, shown as Image 2 in Figure 5. The CT scan may have been taken at a second CT scanning appointment. The second appointment is at a time after the first appointment. In this first example the second appointment is 2 months after the first appointment. The patient may have undergone treatment for a respiratory condition in the period between the first and second appointments. In other examples the second CT capture may be taken a different time period after the first capture, or even before the first CT scan.

[0146] Again, the volume of each of the lobes in the second CT image is calculated using known techniques. The volume of each lobe is shown in Table 1 and shown in Figure 8. In the left lung, the left superior lobe is 450 ml and the left inferior lobe is 900 ml. In the right lung the right superior lobe is 300 ml, the right middle lobe is 360 ml and the right inferior lobe is 750 ml. The total volume of the lungs of the patient in the second CT scan is 3000 ml. This is a larger volume than the 2000 ml volume of the lungs in the first CT scan.

[0147] The percentage of the total volume of the lungs in each of the lobes is calculated and shown in Figure 9. This percentage data allows direct comparison with the data from the first CT scan. For each lobe, the percentage of each lobe is calculated as a percentage of the total volume of the lungs, 3000 ml. This is calculated for each lobe by dividing the volume of the lobe by the total volume and expressing the result a percentage of the total volume. For example, for Image 2, in the left lung the left superior lobe is 450 ml. 450 ml represents 15% of the total volume of 3000 ml. The percentage of the total volume of contained within each lobe for the second scan is also presented in Table 1.

[0148] In the second scan, in the left lung, the left superior lobe includes 15 % of the total volume and the left inferior lobe includes 30 % of the total volume. In the right lung the right superior lobe contains 10 % of the total volume, the right middle lobe contains 20 % of the total volume and the right inferior lobe includes 25 % of the total volume.

[0149] These percentage data are directly compared with the percentages contained within each lobe in other CT scans of the patient (for example compared to the first scan) to assess whether there is a relative change in the regional volume distribution of the lungs of the patient.

[0150] Figure 10 shows diagrammatically the comparison between the percentage data for each of CT scans. The percentage changes for each lobe are shown in Figure 11. The results of the comparison are also shown in Table 2.

[0151] In the left lung, the volume of the left superior lung has changed from 200mL to 450mL, an increase of 250mL. Correspondingly, the percentage of the total volume of the left superior lobe has changed from 10 % to 15 %, an increase of 5 %. This increase in percentage represents a change (an increase) in the relative volume of the superior lobe of the left lung. In the second CT scan, the superior lobe of the left lung contained a 5 % higher proportion of the total lung volume than the first CT. This change in relative percentage of volume in the superior lobe represents a change in the regional volume distribution of the lungs (that is, as a percentage, more air is now flowing into the left superior lobe).

[0152] In the left lung, the volume of the left inferior lung has changed from 700mL to 900mL, an increase of 200mL. Correspondingly, the percentage of the total volume of the inferior lobe has changed from 35 % to 30 %, a decrease of 5 %. This lobe (ie the left inferior lobe) highlights the problem that clinicians encounter when comparing two CT scans taken on different days. Specifically, in this situation if the clinician directly (visually) compares the two CT scans they will notice that the left inferior lobe is darker in Image 2, and will conclude (correctly) that there is more air in the left inferior lung (i.e. determined from visualizing the pixel intensity in the image, which directly corresponds to the density of the tissue, and is measured in Hounsfield units). However, the clinician does not have an easy way of determining that, while there is an increase in volume of the left inferior lung in Image 2, there is actually a relative decrease in volume of the left inferior lung (a reduction of 5%).

[0153] In the right lung, the percentage of the total volume of the right superior lobe has changed from 15 % to 10 %, a decrease of 5 %. In the right middle lobe the percentage of the total volume has changed from 18 % to 20 %, an increase of 2 %. In the right inferior lobe the percentage of the total volume has changed from 22 % to 25 %, an increase of 3 %.

[0154] The changes in percentage of the total volume of the lungs in each of the lobes between the first CT scan and the second CT scan is shown in Figure 11 . The changes represent a change in relative regional volume distribution of the lung between the two CT images. The changes confirm that air is distributed differently in the lungs in the second CT image from the first CT image. For example, in the left lung, a greater proportion of the air is contained in the superior lobe in image 2. Providing a summary of the changes in percentage of the total volume of air in the lungs in each lobe (e.g. as shown schematically in Figure 11 , or in tabular form in Table 2) allows the clinician to rapidly and efficiently digest the complicated changes in the redistribution of ventilation that has occurred between the two scans.

[0155] This change in relative regional volume distribution can indicate a change in lung health. The change in relative regional volume distribution can indicate a change in lung function. For example, a patient suffering from a lung disease may have reduced ventilation (ie airflow into) in a specific region of the lungs. In such a situation the relative volume of the region would be low for this patient. However, after successful treatment, it would be anticipated that this region would experience more typical ventilation, and therefore the relative volume of the region would be more typical. In such an instance there would be an increase in the relative volume of the region between the first scan and the second scan, indicating that the treatment was successful.

Table 2.

[0156] In the first example described above, the change in relative regional volume distribution of the lungs between the two CT scans is calculated using the percentage of the total volume of the lungs in each lobe and comparing the percentages in equivalent lobes in the different CT scans. A second process allows comparison of the CT datasets by normalizing the images to a common volume based on overall volume of the lungs. In the following example, the images are normalized for comparison by adjusting the volume of the first CT scan to match the volume of the second CT scan to allow a direct comparison of regional volumes of air to be made.

[0157] The steps of the second example are now described with reference to Figures 12 to 17. This normalization method is now described using the volume distributions in the first CT scan and the second CT scan shown in Figure 6. Referring again to Figures 6 and 8, the total volume contained within the lungs in the first CT scan is 2000 ml and the total volume contained within the lungs in the second CT scan is 3000 ml. In the first CT scan: in the left lung, the left superior lobe is 200 ml and the left inferior lobe is 700 ml; and, in the right lung the right superior lobe is 300 ml, the right middle lobe is 360 ml and the right inferior lobe is 440 ml. The total volume of the lungs of the patient in the first CT scan is 2000 ml.

[0158] In the second CT scan: in the left lung, the left superior lobe is 450 ml and the left inferior lobe is 900 ml; and, in the right lung the right superior lobe is 300 ml, the right middle lobe is 600 ml and the right inferior lobe is 750 ml. The total volume of the lungs of the patient in the second CT scan is 3000 ml.

[0159] The non-normalised images are compared to calculate the change in volume in each lobe between the CT images. Figure 12 visually maps the equivalent regions of the lungs between the two CT scans. Figure 13 is a graphical representation of the change in volume in the lobes of the lungs between the first scan and second scan. The volume in the superior lobe of the left lung has increased from 200 ml to 450 ml, an increase of 250 ml. The volume in the inferior lobe of the left lung has increased from 700 ml to 900 ml, an increase of 200 ml. The volume in the superior lobe of the right lung is unchanged, from 300 ml to 300 ml. The volume in the middle lobe of the right lung has increased from 360 ml to 600 ml, an increase of 240 ml. The volume in the superior lobe of the left lung has increased from 440 ml to 750 ml, an increase of 310 ml. The overall increase in volume between the first scan and the second scan is 1000 ml.

[0160] Referring again to Figures 6 and 8, the total volume contained within the lungs in the first CT scan is 2000 ml and the total volume contained within the lungs in the second CT scan is 3000 ml. In this example, the images are normalized to a volume of 3000 ml. The volume in the first CT image is 2000 ml. In order to transform the first CT scan to represent a volume of 3000 ml, the volume is increased by a factor of 50% (as increasing 2000mL by 50% will make it 3000mL).

[0161] Figure 14 shows the transformation of the data from the first scan (Image 1 , 2000 ml) to Image 1 ’ (3000 ml). The volumes in each lobe are scaled up by 50%. In the left lung, the volume of the left superior lobe is increased from 200 ml to 300 ml and the volume of the left inferior lobe is increased from 700 ml to 1050 ml. In the right lung, the volume of the right superior lobe is increased from 300 ml to 450 ml, the volume of the right middle lobe is increased from 360 ml to 540 ml, and the volume of the right inferior lobe is increased from 440 ml to 660 ml. The data is shown in Table 3.

Table 3.

[0162] Scan 2 already represents a 3000 ml volume and no transformation is required. Figure 15 shows a comparison of the volumes in the various lobes in Image 1 ’ with Image 2.

Alternatively, both scans could be transformed to some other common value that isn’t either the volume of Scan 1 or Scan 2, for example they could both be transformed to 1000 mL and then compared.

[0163] The volume in each lobe is compared and any change in volume between Scan 1’ and Scan 2 is calculated. The results of the comparison are shown in Figure 16. In the left lung, the difference in volume between Scan 1’ and Scan 2 in the superior lobe is +150 ml and the difference in the inferior lobe is -150 ml. In the right lung, the difference in volume between Scan 1’ and Scan 2 in the superior lobe is -150 ml, the difference in volume in the middle lobe is +60 ml and the difference in volume in the inferior lobe is +90 ml. Since both images are normalized at 3000 ml, the overall difference in volume is zero ml. These change in volume in the lobes represent a change in volume distribution of air in the lungs in the different CT images. [0164] The changes in volume in each of the lobes may be represented as a percentage of the overall volume of the lungs of 3000 ml. These percentages are shown in Figure 17. The percentage change in volume of the left lung is +5 % in the left superior lobe and -5 % in the left inferior lobe. The percentage change in volume of the right lung is -5 % in the right superior lobe, +2 % in the right middle lobe and +3 % in the right inferior lobe.

[0165] In Examples 1 and 2 above the absolute volume of the lungs is calculated from the CT image data. Some further systems allow for the comparison without calculating the absolute volume of the lungs. In one example now described regional volume distribution of the CT image data is defined using the number of voxels in the CT images (without converting the number of voxels into an absolute volume in mL or L). A voxel is a regular 3-dimensional unit of a 3-dimensional digital representation. Voxels are commonly used in the visualization and analysis of medical data to represent and display volume. A voxel represents a unit of volume in the 3-dimensional image.

[0166] Referring to Figures 18 and 19, the processor 2130 is configured to calculate the number of voxels in 3-dimensional CT images of the lungs. The processor 2130 analyses the 3- dimensional CT scan data to calculate the total number of voxels in the CT image of the lungs and also the number of voxels in each of the regions of the lungs. Figure 18 shows the number of voxels in each of the five regions in the first CT image data. Figure 19 shows the number of voxels in each of the five regions in the second CT image data.

[0167] Referring to Figure 18 representing the results from a first CT scan, the total number of voxels in the CT image data is 7,000. In the left lung, the superior lobe includes 700 voxels, corresponding to 10% of the total number of voxels in the lung, and the inferior lobe includes 2450 voxels, corresponding to 35% of the total. In the right lung, the superior lobe includes 1050 voxels, corresponding to 15% of the total, the middle lobe includes 1260 voxels, corresponding to 18% of the total, and the inferior lobe includes 1540 voxels, corresponding to 22 % of the total.

[0168] Referring to Figure 19 representing the results from a second CT scan, the total number of voxels in the CT image data is 10,500. In the left lung, the left superior lobe includes 1575 voxels, corresponding to 15% of the total number of voxels in the lung, and the left inferior lobe includes 3150 voxels, corresponding to 30% of the total. In the right lung, the right superior lobe includes 1050 voxels, corresponding to 10% of the total, the right middle lobe includes 2100 voxels, corresponding to 20% of the total, and the right inferior lobe includes 2625 voxels, corresponding to 25% of the total.

[0169] Since both CT image data sets are now represented in terms of percentage of voxels in each region, they can be directly compared to determine any changes in the relative regional volume between the CT scans for each of the regions. Figure 20 shows the change in relative regional volume between image 1 and image 2 based on data obtained by voxels. This result is the same as that calculated using absolute volume measurements.

[0170] A further example of calculating the change in volume distribution of the lungs between two CT scans based on calculations of total volume of the lungs is now described with reference to Figures 22 to 27. In this example, the inferior lobe of the right lung is selected.

[0171] Figure 22 is a schematic figure representing the volume of the lung calculated from a first CT scan on a patient. The CT scan is taken during an inhalation breath hold. Volume is calculated from the CT scan using techniques described above. In the example, the number of voxels in the lung and the number of voxels in the right inferior lobe, as measured in the CT image data, is used to calculate volume. In the example of Figure 22, the volume of the right inferior lobe is 500 ml. The total volume of the lungs of the patient in the first CT scan of Figure 22 is 2000 ml. The relative volume of the right inferior lobe is calculated as a proportion of the volume of the entire lungs. The selected region, i.e. the right inferior lobe, includes 500 ml of the entire lung volume of 2000 ml. This represents 25 % of the volume of the entire lungs, as represented in Figure 23.

[0172] Figure 24 is a schematic figure representing the volume of the lungs calculated from a second CT scan on the patient. The second CT scan is taken at a time after the first CT scan. The CT scan is again taken during an inhalation breath hold. Volume is calculated from the CT scan using techniques described above. In the scan represented in Figure 24, the volume of the right inferior lobe is calculated to be 600 ml. The total volume of the lungs of the patient in the second CT scan of Figure 24 is calculated to be 3000 ml. The relative volume of the right inferior lobe is calculated as a proportion of the volume of the entire lungs. The right inferior lobe includes 600 ml of the entire lung volume of 3000 ml. This represents 20 % of the volume of the entire lungs, as represented in Figure 25.

[0173] Despite the right inferior lobe having a greater volume in the second scan when compared with the first scan (i.e. 600 ml compared with 500 ml), in the second scan the right inferior lobe actually has a lower proportional volume than in the first scan (i.e. 20 % of total volume of the lungs compared with 25 %), as shown in Figure 26, i.e. a reduction of 5 % in the relative regional volume of the right inferior lobe (as shown in Figure 27). This change in relative regional volume of the right inferior lobe represents a change in volume distribution within the lung.

[0174] The changes in percentage of the total volume of air in the lungs in each of the lobes between the first CT scan and the second CT scan is shown in Figure 27.

[0175] The changes represent a change in relative regional volume distribution of air in the lung between the two CT images. The change confirms that volume of the lung is distributed differently in the lungs in the second CT image from the first CT image. Typically, this change in relative regional volume in the right inferior lobe is due to a change in the relative regional volume of air in the right inferior lobe during the scans. In the second scan, a smaller proportion of the total volume of the lung is in the right inferior lobe. This change in relative regional volume in the right inferior lobe is an indication that air is being distributed in the lung differently in the second scan than in the first scan. This redistribution may be an indication of a change in lung health and/or a change in lung function.

[0176] The example shows how changes in lung health can be identified by comparing the relative regional volume of a single region of the lung. In other embodiments, the relative regional volume of multiple regions of the lung can be calculated to identify changes in lung health. In other embodiments the relative regional volume of all regions of the lung may be calculated to identify changes in lung health.

[0177] In some of the examples the relative volume of the region is calculated with respect to the volume of the entire lungs. In other embodiments, the relative volume of the region of the lung is calculated with respect to a comparative larger part of the lung. The comparative larger part of the lung may be the left lung, the right lung, or the entire lungs. Or the comparative larger part of the lung may be a selected portion of the lung that does not correspond exactly to the recognised anatomical parts of the lung of the left lung, the right lung or the entire lungs, for example it may be a part of the left lung. The relative volume of the region is calculated by calculating the volume of the region and calculating the volume of the comparative larger part of the lung. The relative volume of the region expresses the volume of the region with respect to the volume of the comparative larger part. This may be calculated by dividing the volume of the region by the volume of the comparative larger part of the lung. The relative volume is unitless and may be expressed as a fraction, or percentage of the volume of the comparative larger part of the lung. The relative volume of the region is calculated for different breaths.

[0178] In further examples, transformation data between the first CT image dataset and the second CT image dataset may be used to directly assess the change in relative regional distribution of air in the lung. The example mapping algorithms described above, namely Deformable Image Registration (DIR) and Particle Image Velocity (PIV). These mapping processes obtain transformation data required to morph one dataset onto the other, and thus to map region of interest to be investigated to be mapped to the corresponding region in the other CT scan. This transformation data can be used to obtain regional change data which provides data defining a change in relative regional distribution of air in the lung between the image datasets. This data may be obtained on a voxel by voxel basis allowing the change in relative regional volume in the lung to be calculated at a high resolution.

[0179] In some of the examples above, analysis of the CT images of the lungs is performed by dividing the lungs into five regions corresponding to the lobes of the lungs, analyzing each region in each image and then comparing the corresponding regions in each image to assess any differences. Both CT images used in the comparison are divided using the same regional division.

[0180] In further embodiments analysis may be performed by using a different technique for dividing the images of the lungs into regions. The images may be divided based on anatomical structure of the lung, for example based on individual voxels, groups of voxels or sub-lobes. Alternatively, the images may be divided geometrically. Other arbitrary techniques for dividing the images may be used. Examples of other techniques include geometric-type techniques dividing the images by blocks or curves. The images to be compared are divided using the same regional division.

[0181] In the examples above the equivalent regions of the images of the lungs are compared to identify a relative regional change in the distribution of air in the lungs between the images.

[0182] In one further example embodiment, changes in relative regional volume of the lung are assessed by comparing the relative shapes of different regions of the lungs. By comparing the relative shapes of the different regions between different scans an assessment of any relative regional changes in shape can be performed. Any change in relative regional shape may be an indication of a change in lung health. [0183] In one further example embodiment, changes in the relative expansion of different regions of the lungs can be calculated. By comparing the relative expansion of the different regions between different scans an assessment of any relative regional changes in expansion can be performed. Any change in relative regional expansion may be an indication of a change in lung health.

[0184] In one further example embodiment, changes in relative size of the lung are assessed by comparing the relative size of different regions of the lungs. By comparing the relative size of the different regions between different scans an assessment of any relative regional changes in size can be performed. Any change in relative regional size may be an indication of a change in lung health.

[0185] One of the benefits of the various embodiments is that clinicians are able to compare CT image datasets of patients taken at different times, on different equipment, or under different conditions to assess whether the patient has experienced a change in lung health.

Embodiments also allow images to be compared which were captured with different volumes of air within the lungs. One of the exciting benefits is that if patients already have a number of CT scans on their file dating back in time, a clinician is able to look back at those scans and compare them with more recent scans to assess any change in lung health over time. This allows analysis and diagnosis to be performed retrospectively.

[0186] A core measurement is how the distribution of air within the lung has changed over time between scans. This change in distribution of air can provide important information about change in lung function and lung health.

[0187] The term ‘image’ or ‘images’ as used herein may correspond to image data or image datasets from which a visual image may be created. While these image datasets may be converted to visual images, it is understood that the processing of the images by the system is usually with respect to the datasets. The steps of manipulating and comparing the images described above, including for example deforming images, normalizing images, assessing images and comparing images, may be performed on data representing the images. These steps may involve the representation of the dataset as a visual image (either digital, physical or otherwise). The steps are usually performed on the datasets. These data may be manipulated, analysed and compared within a computing system or other environment to identify changes in lung health of the patient without requiring the creation of visual images, whether digital, physical or otherwise.

[0188] It is to be understood that, if any prior art publication is referred to herein, such reference does not constitute an admission that the publication forms a part of the common general knowledge in the art, in Australia or any other country.

[0189] In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as "comprises" or "comprising" is used in an inclusive sense, namely, to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

[0190] It is to be understood that the aforegoing description refers merely to preferred embodiments of invention, and that variations and modifications will be possible thereto without departing from the spirit and scope of the invention, the ambit of which is to be determined from the following claims.

Claims

CLAIMS:

1 . A method of identifying changes in lung health using 3-dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath; calculating a relative regional volume of a region of the lung for each of the two 3-dimensional image datasets; comparing the relative regional volumes calculated at each breath to assess a change in relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

2. A method according to claim 1 wherein the step of calculating a relative regional volume of a region of the lung is calculated with respect to a comparative larger part of the lung.

3. A method according to claim 2 wherein the comparative larger part of the lung is one of: the left lung; the right lung; or, the combination of the left lung and the right lung.

4. A method according to any one of claims 2 or 3 wherein the relative regional volume of the region of the lung is expressed as a proportion, fraction or a percentage of the volume of the comparative larger part of the lung.

5. A method of identifying changes in lung health using 3-dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath of a lung; calculating a relative regional distribution of air in the lung for each of the two 3-dimensional image datasets; comparing the relative regional distributions of air in the lung calculated at each breath to assess a change in relative regional distribution of air in the lung between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

6. A method according to claim 5 wherein the step of comparing the relative regional distribution of air in the lung comprises the steps of determining the total volume of air in the lung for each breath, normalizing the image datasets to a common total volume of air, and comparing the normalized image datasets by comparing the volume of air in equivalent regions of the normalized image datasets.

7. The method according to any one of claims 5 or 6 wherein the step of calculating a relative regional distribution of air in the lung for each of the two 3-dimensional image datasets comprises the step of dividing the 3-dimensional image datasets into at least two lung regions and calculating the regional distribution of air in the lung for each of the at least two regions.

8. A method according to any one of claims 1 to 7 wherein the region of the lung is one of the superior lobe of the left lung, the inferior lobe of the left lung, the superior lobe of the right lung, the middle lobe of the right lung and the inferior lobe of the right lung.

9. A method according to any one of claims 1 to 8 wherein the lungs are divided into one of: more than five regions, at least 5 regions; at least 6 regions; at least 10 regions; at least 15 regions; at least 18 regions; at least 19 regions; at least 20 regions; at least 25 regions; at least 40 regions; at least 50 regions; or, at least 100 regions.

10. A method according to any one of claims 1 to 9 wherein the region is defined by at least one voxel in a 3-dimensional image of the lung.

11 . A method according to any one of claims 1 to 10 wherein each of the 3-dimensional images of the lungs is acquired using computerized tomography (CT) technique.

12. A method according to any one of claims 1 to 11 wherein at least one of the images is acquired at an inspiration breath hold.

13. A method according to any one of claims 1 to 12, the method being performed without requiring a visual image, whether digital, physical or otherwise, to be produced.

14. A method of identifying changes in lung health using 3-dimensional image datasets, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquiring two 3-dimensional image datasets of a lung, each image dataset being acquired during a different breath of a lung; comparing the image datasets to assess a change in relative regional distribution of air in the lungs between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

15. A system for identifying changes in lung health using 3-dimensional image datasets, including: an image acquisition module, configured to acquire two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath; a processor configured to calculate a relative regional volume of a region of the lung for each of the two 3-dimensional image datasets; and compare the relative regional volumes calculated at each breath to assess a change in relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

16. A non-transitory computer readable storage medium having a computer program stored therein, that when executed by a processor of a computer, causes the computer to execute steps directed to identifying changes in lung health using 3-dimensional image datasets, including: acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath; calculating a relative regional volume of a region of the lung for each of the two 3- dimensional image datasets; comparing the relative regional volumes calculated at each breath to assess a change in relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

17. A method of identifying changes in lung health using 3-dimensional images, including acquiring two 3-dimensional image datasets of a lung, each 3-dimensional image dataset being acquired during a different breath of a lung; normalizing the two three-dimensional image datasets to produce two normalized 3-dimensional image datasets, comparing the two normalized 3-dimensional image datasets to assess a change in relative regional distribution of air in the lung between the two breaths, wherein the change in relative regional distribution of air in the lung is indicative of a change in lung health.

18. A method of identifying changes in lung health using two 3-dimensional image datasets, including: acquiring a first 3-dimensional image dataset of a lung and a second 3-dimensional image dataset of the lung, the second 3-dimensional image dataset being acquired in a subsequent breath to the first 3-dimensional image dataset; defining a lung tissue region in one of the first and second 3-dimensional image datasets, the lung tissue region consisting of one or more voxels in the 3-dimensional image dataset; mapping the lung tissue region to the corresponding lung tissue in the other 3- dimensional image dataset to define a corresponding lung tissue region; normalizing the volume of the lung tissue region to the volume of the 3-dimensional image dataset it was derived from, and normalizing the volume of the corresponding lung tissue region to the volume of the 3-dimensional image dataset it was derived from, to create two relative regional volumes; comparing the two relative regional volumes to calculate a change in the relative regional volume of the region between the two breaths, wherein the change in relative regional volume is indicative of a change in lung health.

19. A method of identifying changes in lung health using two 3-dimensional image datasets, including: acquiring a first 3-dimensional image dataset of a lung in a first breath, and a second 3- dimensional image dataset of the lung in a second breath, the first breath and the second breath being different breaths; defining a lung tissue region in the first 3-dimensional image dataset, the lung tissue region consisting of one or more voxels in the 3-dimensional image dataset; mapping the lung tissue region in the first 3-dimensional image dataset to the corresponding lung tissue in the second 3-dimensional image dataset to define a corresponding lung tissue region in the second 3-dimensional image dataset; normalizing the volume of the lung tissue region in the first 3-dimensional image dataset to the volume of the first 3-dimensional image dataset to create a first regional volume distribution measurement; normalizing the volume of the corresponding lung tissue region in the second 3- dimensional image dataset to the volume of the second 3-dimensional image dataset to create a second regional volume distribution measurement; comparing the first regional volume distribution measurement to the second regional volume distribution measurement to calculate a change in the regional volume distribution measurement between the two breaths, wherein the change in the regional volume distribution measurement is indicative of a change in lung health.