CN110246144A - A kind of the blood vessel Enhancement Method and system of lung CT image - Google Patents

Abstract

The present invention relates to a blood vessel enhancement method and system for lung CT images, including a CT image input unit, a preset segmentation threshold acquisition unit, a lung tissue area acquisition unit without blood vessels, a blood vessel threshold calculation unit, an initial diffusion surface acquisition unit, According to the initial diffusion surface and the blood vessel threshold, surface diffusion is performed in the closed lung tissue area with blood vessels in the n-layer chest CT image to obtain pulmonary blood vessels. It can automatically and accurately extract pulmonary blood vessels from chest CT images, making doctors' observation of pulmonary blood vessels more intuitive and accurate, avoiding the interference of the heart and bones when observing and diagnosing pulmonary vascular diseases on images, and the calculation speed is fast. short.

Description

A blood vessel enhancement method and system for lung CT images

technical field

The invention relates to the technical field of diagnosis and treatment of lung cancer, in particular to a blood vessel enhancement method and system for lung CT images.

Background technique

CT is the abbreviation of computer X-ray tomography technology, which is a commonly used medical imaging equipment. CT images are black and white images, with different gray levels representing the degree of absorption of X-rays by organs and tissues. For example, in a chest CT image, areas with low density (that is, low gray value) represent trachea and lung parenchyma, and areas with high density (that is, high gray value) represent blood vessels, chest cavity, bones, etc. CT can intuitively display the lesion area on the image, which provides a convenient and reliable basis for doctors to observe and diagnose diseases. Usually, CT images are cross-sectional images, and in order to display the entire organ, multiple consecutive slice images are required.

In the CT diagnosis of lung diseases, it is usually necessary to scan the entire thoracic cavity to obtain pulmonary vascular images. In the scanned images, there must be bones and hearts in the thoracic cavity, and clinicians can only check the lesion area through layer-by-layer image observation. . Therefore, accurate extraction of pulmonary vascular tissue from CT images is an effective method to eliminate the interference of bone, heart and other tissues, so that doctors can observe and diagnose pulmonary vascular diseases more intuitively.

Contents of the invention

The technical problem to be solved by the present invention is to provide a blood vessel enhancement method and system for lung CT images, so as to solve the defects in the prior art.

The technical scheme that the present invention solves the problems of the technologies described above is as follows:

A device for extracting pulmonary vessels from a chest CT image, comprising: a CT image input unit: used to receive an input CT image; a preset segmentation threshold acquisition unit: using the average projection density of the three-dimensional features of the CT image combined with multi-scale dot enhancement to perform prediction Processing; lung tissue area acquisition unit without blood vessels: used to select specified pixel points of the lung area on the specified image layer to obtain initial seed points, according to the preset segmentation threshold and initial seed points, in the n Perform 3D region growth on the chest CT image to obtain the lung tissue area without blood vessels; blood vessel threshold calculation unit: perform morphological operations on the lung tissue area to obtain the closed lung tissue area with blood vessels, and calculate the blood vessel threshold; initial diffusion Surface acquisition unit: used to search the closed lung tissue area with blood vessels on the specified image layer to find the pixel points larger than the blood vessel threshold as initial marker points, and obtain the initial diffusion surface; pulmonary blood vessel extraction unit: used to obtain the initial diffusion surface according to the initial diffusion surface and blood vessel thresholding, performing surface diffusion in the closed lung tissue area with blood vessels in the n-slice chest CT image to obtain pulmonary blood vessels.

A method for extracting pulmonary vessels from chest CT images, comprising the steps of:

The first step: preprocessing by using the average projection density of CT image 3D features combined with multi-scale dot enhancement;

Step 2: select specified pixel points in the lung region on the specified image layer to obtain initial seed points, and perform 3D region growth on the n-layer chest CT images according to the preset segmentation threshold and initial seed points to obtain Areas of lung tissue without blood vessels;

Step 3: Perform morphological calculations on the lung tissue area to obtain a closed lung tissue area with blood vessels, and calculate the blood vessel threshold;

Step 4: Using the DBSCAN superpixel sequence image clustering algorithm optimized by the strategy of clustering only pulmonary nodules and the fast and accurate adaptive threshold, cluster the superpixel samples to obtain the sequential pulmonary nodule mask, Finally, the sequential lung nodule image of lung CT is obtained;

Step 5: According to the initial diffusion surface and the blood vessel threshold, surface diffusion is performed in the closed lung tissue area with blood vessels in the n-layer chest CT image to obtain pulmonary blood vessels.

The beneficial effects of the present invention are: the pulmonary vessels can be automatically and accurately extracted from the chest CT image, making the doctor's observation of the pulmonary vessels more intuitive and accurate, and avoiding the heart and bones when observing and diagnosing pulmonary vascular diseases on the images. Interference, and the calculation speed is fast and the time is short.

Description of drawings

Fig. 1 is a structural representation of the present invention;

Detailed ways

The principles and features of the present invention are described below in conjunction with the accompanying drawings, and the examples given are only used to explain the present invention, and are not intended to limit the scope of the present invention.

As shown in Figure 1, a blood vessel enhancement method and system for lung CT images, including

A device for extracting pulmonary vessels from a chest CT image, comprising: a CT image input unit: used to receive an input CT image; a preset segmentation threshold acquisition unit: using the average projection density of the three-dimensional features of the CT image combined with multi-scale dot enhancement to perform prediction Processing; lung tissue area acquisition unit without blood vessels: used to select specified pixel points of the lung area on the specified image layer to obtain initial seed points, according to the preset segmentation threshold and initial seed points, in the n Perform 3D region growth on the chest CT image to obtain the lung tissue area without blood vessels; blood vessel threshold calculation unit: perform morphological operations on the lung tissue area to obtain the closed lung tissue area with blood vessels, and calculate the blood vessel threshold; initial diffusion Surface acquisition unit: used to search the closed lung tissue area with blood vessels on the specified image layer to find the pixel points larger than the blood vessel threshold as initial marker points, and obtain the initial diffusion surface; pulmonary blood vessel extraction unit: used to obtain the initial diffusion surface according to the initial diffusion surface and blood vessel thresholding, performing surface diffusion in the closed lung tissue area with blood vessels in the n-slice chest CT image to obtain pulmonary blood vessels.

A method for extracting pulmonary vessels from chest CT images, comprising the steps of:

The first step: preprocessing by using the average projection density of CT image 3D features combined with multi-scale dot enhancement;

Step 2: select specified pixel points in the lung region on the specified image layer to obtain initial seed points, and perform 3D region growth on the n-layer chest CT images according to the preset segmentation threshold and initial seed points to obtain Areas of lung tissue without blood vessels;

Step 3: Perform morphological calculations on the lung tissue area to obtain a closed lung tissue area with blood vessels, and calculate the blood vessel threshold;

Step 4: Using the DBSCAN superpixel sequence image clustering algorithm optimized by the strategy of clustering only pulmonary nodules and the fast and accurate adaptive threshold, cluster the superpixel samples to obtain the sequential pulmonary nodule mask, Finally, the sequential lung nodule image of lung CT is obtained;

Step 5: According to the initial diffusion surface and the blood vessel threshold, surface diffusion is performed in the closed lung tissue area with blood vessels in the n-layer chest CT image to obtain pulmonary blood vessels.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within range.

Claims (2)

1. a blood vessel enhancement method of lung CT image, is characterized in that, comprises the steps: The first step: preprocessing by using the average projection density of CT image 3D features combined with multi-scale dot enhancement; Step 2: select specified pixel points in the lung region on the specified image layer to obtain initial seed points, and perform 3D region growth on the n-layer chest CT images according to the preset segmentation threshold and initial seed points to obtain Areas of lung tissue without blood vessels; Step 3: Perform morphological calculations on the lung tissue area to obtain a closed lung tissue area with blood vessels, and calculate the blood vessel threshold; Step 4: Using the DBSCAN superpixel sequence image clustering algorithm optimized by the strategy of clustering only pulmonary nodules and the fast and accurate adaptive threshold, cluster the superpixel samples to obtain the sequential pulmonary nodule mask, Finally, the sequential lung nodule image of lung CT is obtained; Step 5: According to the initial diffusion surface and the blood vessel threshold, surface diffusion is performed in the closed lung tissue area with blood vessels in the n-layer chest CT image to obtain pulmonary blood vessels. 2. A blood vessel enhancement system for lung CT images, characterized in that it includes: a CT image input unit: used to receive input CT images; a preset segmentation threshold acquisition unit: using the CT image three-dimensional feature average projection density combined with a multi-scale circle Point enhancement for preprocessing; lung tissue area acquisition unit without blood vessels: used to select specified pixel points of the lung area on the specified image layer to obtain initial seed points, according to the preset segmentation threshold and initial seed points, Perform 3D region growth on the n-layer chest CT image to obtain a lung tissue area without blood vessels; the blood vessel threshold calculation unit: perform morphological operations on the lung tissue area to obtain a closed lung tissue area with blood vessels, and calculate blood vessels Threshold; initial diffusion surface acquisition unit: used to find pixels larger than the blood vessel threshold in the closed lung tissue area with blood vessels on the specified image layer as the initial marker point, and obtain the initial diffusion surface; pulmonary blood vessel extraction unit: used to obtain the initial diffusion surface according to For the initial diffusion surface and the blood vessel threshold, surface diffusion is performed in the closed lung tissue area with blood vessels in the n-layer chest CT image to obtain pulmonary blood vessels.