[go: up one dir, main page]

WO2021233795A1 - Directives de décision de radiologie personnalisées tirées d'une imagerie analogique passée et d'un phénotype clinique applicables au point de lecture - Google Patents

Directives de décision de radiologie personnalisées tirées d'une imagerie analogique passée et d'un phénotype clinique applicables au point de lecture Download PDF

Info

Publication number
WO2021233795A1
WO2021233795A1 PCT/EP2021/062906 EP2021062906W WO2021233795A1 WO 2021233795 A1 WO2021233795 A1 WO 2021233795A1 EP 2021062906 W EP2021062906 W EP 2021062906W WO 2021233795 A1 WO2021233795 A1 WO 2021233795A1
Authority
WO
WIPO (PCT)
Prior art keywords
radiology
patient
patient data
examination
cohort
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/EP2021/062906
Other languages
English (en)
Inventor
Abhivyakti SAWARKAR
Vadiraj Krishnamurthy HOMBAL
Thomas Buelow
Sandeep Madhukar Dalal
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Koninklijke Philips NV
Original Assignee
Koninklijke Philips NV
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips NV filed Critical Koninklijke Philips NV
Publication of WO2021233795A1 publication Critical patent/WO2021233795A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients

Definitions

  • the following relates generally to the imaging arts, radiology arts, radiology reading arts, radiomics arts, and related arts.
  • ACR American College of Radiology
  • AC Appropriateness Criteria
  • practice-based evidence may comprise of records of patients with a similar diagnosis, demographics, image phenotype, image reports, as well as treatments, responses, side effects, complications, and clinical outcomes.
  • This aggregate data set of clinical care and outcomes represents local practices and experiences, and contains valuable retrospective observational data that can be re-purposed as a personalized evidence base to make informed decisions for the patient at hand.
  • observational practice-based evidence may be the best data available for point of care decision making and more robust than expert recollection, consensus derived guidelines or opinion alone.
  • ACR develops screening, safety, communication, appropriateness guidelines that standardize the practice of radiology.
  • ACR-AC are the most comprehensive evidence based guidelines for diagnostic imaging selection and image guided interventional procedures, by embodying the current evidence for selecting appropriate diagnostic imaging and interventional procedures for numerous clinical conditions. Evidence from expert consensus, literature and other sources, studies and clinical trials typically has limitations, rests on restricted patient-inclusion criteria, and does not generalize well to real clinical situations.
  • Physicians see many patients on a daily basis that do not have symptoms well-aligned with the applicable clinical guidelines and need more personalized care direction than those offered by existing guidelines or clinical acumen. Since the large scale adoption of electronic health records, an extraordinary amount of patient data is generated, potentially enabling practice-based evidence, in which data is re-purposed to benefit current and future patients in a learning healthcare system.
  • Radiomics Imaging phenotype analyses using radiomics have been shown to have expediency in prognosis, non-invasive disease tracking, and evaluation of disease response to treatment.
  • Radiomics denotes the extraction and analysis of large amounts of advanced quantitative imaging features reflecting radiologic spatial distributions within a lesion, and it has been expected to provide descriptive and predictive models linking image features to lesion phenotypes.
  • radiomics has generally not been effectively leveraged in the radiology reading environment, beyond the extent that a given radiologist may recognize relevance of a published clinical study employing radiomics to a particular radiology examination being read.
  • a radiology reading system comprises a radiology workstation including a display device and at least one user input device, and at least one electronic processor operatively connected with the display device and the at least one user input device and programmed to: open a radiology examination of a patient; provide an image rendering engine configured to render radiological images of the opened radiology examination on the display device of the radiology workstation; and provide a radiology report entry user interface for receiving a radiology report on the opened radiology examination via the at least one user input device of the radiology workstation and for displaying a report window showing at least a portion of the received radiology report on the display device of the radiology workstation.
  • the at least one electronic processor is further programmed to: extract patient features from one or more electronic patient databases for the patient of the opened radiology examination; extract examination features for the opened radiology examination; performing a patient cluster analysis including a radiomics analysis including at least determining a cohort of similar patients having patient records stored in the one or more electronic patient databases based on the extracted patient features and the extracted examination features; perform at least one patient data analysis and radiomics analysis on the cohort; and concurrently with the providing of the radiology report entry user interface, further providing a patient data analytics user interface (UI) including a patient data analytics window displaying at least one result of the at least one patient data analysis on the display device of the radiology workstation.
  • UI patient data analytics user interface
  • a non-transitory computer readable medium stores instructions executable by at least one electronic processor to perform a method of using patient cluster analytics and radiomics in a medical examination reading environment.
  • the method includes: providing a radiology reading environment on a radiology workstation that includes or is operatively connected with the at least one electronic processor including opening a radiology examination of a patient; and while the radiology examination of the patient is opened in the radiology reading environment, providing a patient data analytics analysis including: extracting patient features from patient data of the patient of the opened radiology examination; extracting examination features for the opened radiology examination; determining a cohort of similar patients based on the extracted patient features and the extracted examination features; and providing a patient data analytics UI component of the radiology reading environment showing information about the cohort on the radiology workstation.
  • a method of using similar past patient cluster analytics including radiomics in a radiology reading environment includes: opening a radiology examination of a patient; providing an image rendering engine configured to render radiological images of the opened radiology examination on the display device of the radiology workstation; providing a radiology report entry user interface for receiving a radiology report on the opened radiology examination via the at least one user input device of the radiology workstation and for displaying a report window showing at least a portion of the received radiology report on the display of the radiology workstation; performing a similar past patient cluster analytics including radiomics analysis including: extracting patient features from one or more electronic patient databases for the patient of the opened radiology examination; extracting examination features for the opened radiology examination; determining a cohort of similar patients having patient records stored in the one or more electronic patient databases based on the extracted patient features and the extracted examination features; concurrently with the providing of the radiology report entry user interface, providing a patient data analytics user interface including a patient data analytics window displaying at least one result of
  • Another advantage resides in automatically searching a clinical database and/or a radiology information system database to identify data and images from cohorts of similar patients to a patient undergoing a medical examination.
  • Another advantage resides in providing a radiology workstation including a dashboard containing aggregated data from cohorts of analogous patients, which further provides the radiologist with dynamic access to analogous patient cohort analyses performed in real-time for specific phenotypes presented in a radiology examination currently undergoing reading.
  • Another advantage resides in providing aggregate data at the point of reading from cohorts of similar patients to the imaged patient, to provide information to diagnose the imaged patient.
  • Another advantage resides in using data and images from cohorts of similar patients, in conjunction with data of an imaged patient, to diagnose the imaged patient.
  • Another advantage resides in providing a radiology workstation configured to efficiently apply data analysis from cohort of analogous patients operating on content of one or more electronic patient databases and tailored to phenotype information derived from a currently read radiology examination to diagnose a patient, thereby saving time and money while improving diagnosis.
  • Another advantage resides in providing a dashboard of analytics from cohorts of analogous patients on the radiology reading workstation.
  • Another advantage resides in providing a dashboard, at the point of reading, of cohort of patients with similar imaged phenotype to a current patient, and similar patient data to the current patient (e.g. similar demographics, and/or similar chronic conditions, et cetera) selected and refined algorithmically and to provide the radiologist with access to patient data of the cohort for consideration during the reading of the radiology examination of the current patient.
  • similar patient data e.g. similar demographics, and/or similar chronic conditions, et cetera
  • Another advantage resides in providing the radiology reading workstation with a chatbot user interface to allow a user to input specific patient selection criteria in order to refine analytics from the cohort of patients with similar demographic and/or imaging data to a patient whose radiology examination is undergoing reading.
  • a given embodiment may provide none, one, two, more, or all of the foregoing advantages, and/or may provide other advantages as will become apparent to one of ordinary skill in the art upon reading and understanding the present disclosure.
  • FIGEIRE 1 diagrammatically illustrates an illustrative radiology apparatus in accordance with the present disclosure.
  • FIGEIRE 2 shows example flowchart operations performed by the apparatus of
  • FIGURE 1 A first figure.
  • FIGURE 3 shows an example of an output displayed on the apparatus of
  • FIGURE 1 A first figure.
  • EHR electronic health record
  • Radiomics operations can be implemented. Radiomics involves extraction and analyzes of large amounts of advanced quantitative imaging features at cohort or population levels to provide models relating image features to lesion phenotypes.
  • the potential of evidence from similar past patients and radiomics for patient diagnosis and assessment has been recognized in clinical study settings. However, it has not been well integrated into the radiology reading environment. In existing radiology workflow, there is no practical way to apply applicable practice evidence drawn from similar past patients including radiomics to an individual radiology examination reading. While the radiologist might consult published practice guidelines and scientific papers presenting relevant clinical precedence, such findings were usually developed for a restricted population cohort not closely aligned or personalized for the specific patient whose examination is being read.
  • the patient cohorts in published radiology papers are often mismatched with the patient whose radiology examination is being read in terms of demographics, image modality or specifics of the imaging scans, image phenotype, or so forth.
  • a radiologist is usually expected to complete a radiology reading in a few minutes to a few tens of minutes, which does not provide sufficient time look for personalized applicable diagnostic guidelines from past analogous patients with a comparable phenotype.
  • the disclosed system Upon loading an imaging examination into the radiology workstation, the disclosed system automatically reads the current patient data, and then searches past patient information from patient database (e.g., an Electronic Medical Record (EMR), a Picture Archiving and communication System (PACS), a Radiology Information System (RIS) database, etc.) and extracts patient features such as patient demographic data (age, gender, ethnicity, etc.), patient symptoms, diagnoses (if known), medications, laboratory test results, and/or so forth.
  • patient database(s) leveraged are geographically local to the point of reading, for example including the electronic medical record (EMR) of the hospital hosting the radiology department at which the reading is being performed.
  • these local database(s) are likely to contain patient records for patients who are similarly situated to the patient whose examination is being read. Additionally, the radiology examinations of these similarly situated patients may have been acquired at the same radiology department using the imaging devices and imaging scan protocols employed by that local radiology department. These factors significantly enhance the likelihood that analyses of these similarly situated patients may yield information that is highly relevant to the reading of the current radiology examination.
  • the patient searching may be expanded, upon request by the radiologist, to a database of larger geographic extent (e.g., the records of an entire network of mutually affiliated hospitals) to provide a larger cohort size.
  • the disclosed system automatically searches one or more clinical database(s) for a cohort of similar patients who have had relevant radiology examinations (e.g., of the same imaged anatomy using the same modality).
  • relevant radiology examinations e.g., of the same imaged anatomy using the same modality.
  • the disclosed system also addresses the issue that the radiologist has severe time constraints, excess information, and difficulty in accessing relevant clinical insights in performing the reading and can benefit from targeted and relevant clinical information provided readily at the point of reading.
  • the radiologist is expected to complete a reading in a timeframe of a few minutes to a few tens of minutes.
  • the disclosed system accommodates such time constraint by providing easy access to the relevant data insights drawn from analogous past patient clusters, in addition to radiomic data generated by the system, but without interfering with the reading process.
  • An icon is provided on the radiology workstation in some embodiments which, if selected, brings up a dashboard with information about the cohort.
  • the dashboard also provides for review of images of the cohort patients, e.g. as a browsable stack of image grid thumbnails via which the radiologist can select an image grid of a specific patient for more detailed review (and, can then select specific images of the grid for more detailed review).
  • the radiologist can also select to review other portions of the selected patient’s medical record.
  • the system provides the user with an input via which the radiologist can refine the definition of the cohort.
  • the radiologist can enter a specific radiology finding into the input, and the system will return the sub-cohort of the cohort patients whose radiology examinations included that finding.
  • the radiologist may expand the cohort by removing one or more parameters defining the cohort. For example, if the initial automatically selected cohort was restricted to male patients in the database(s), the radiologist may choose to remove the gender constraint so that the cohort is expanded to include similarly situated female patients in the database(s).
  • the input in some embodiments is a search engine input, i.e. a freeform text input via which the radiologist enters one or more keywords with optional Boolean logic such as “AND” or “OR” connectors.
  • a conversational chatbot is provided via which the user can enter questions in natural language, and the chatbot extracts the relevant clinical terms and refines (or expands) the (sub-)cohort accordingly.
  • the chatbot approach advantageously can link a series of questions.
  • the user might ask for the patients with a first finding to produce a first sub-cohort, then ask about a second finding to produce a second sub-cohort, and then ask about patients having both findings to produce a third sub-cohort which is the intersection of the first and second sub-cohorts.
  • the disclosed systems and methods integrate personalized, applicable practice based evidence derived from cluster of past analogous patients who have a comparable clinical and imaging phenotype, which among other methods also uses radiomics, into the radiology reading environment.
  • the radiology reading system 10 includes, or is used in conjunction with, a networked computing system 12.
  • the networked computing system 12 may comprise a single server computer, a computing cluster, a cloud computing resource, or so forth.
  • One or more databases such as a PACS database 14, a EMR (also referred to herein as an HER) database 15, and a RIS database 16, are installed on the networked computing system 12, and are connected.
  • FIGURE 1 illustrates a single representative radiology electronic processing device, such as a radiology workstation 18 (or more generally, a computer), via a secure electronic data network, such as a wired and/or wireless Wide Area Network (WAN) implemented via Ethernet, Wi-Fi, or another suitable wired and/or wireless electronic data networking protocol.
  • WAN Wide Area Network
  • the secure electronic data network should have sufficient bandwidth to communicate radiology images, which are typically large data files, to and from the radiology workstation 18.
  • the databases 14, 15, 16, installed on the networked computing system 12 may be connected with other computing systems such as physician’s desktop computers, radiological imaging system controllers (e.g.
  • PACS Picture Archiving and Communication System
  • EMR Electronic Medical Record
  • EHR Electronic Health Record
  • the patient record in the EMR or EHR may include hyperlinks to radiology examinations stored in the PACS
  • the PACS record for a patient may include a hyperlink to the patient’s record in the EMR or EHR.
  • the PACS may incorporate RIS content, forming an integrated RIS/PACS database.
  • the PACS stores radiology images in accordance with the Digital Imaging and Communications in Medicine (DICOM) file format definition promulgated by the National Electrical Manufacturers Association (NEMA), or in a variant of the standard DICOM definition.
  • DICOM Digital Imaging and Communications in Medicine
  • the radiology workstation 18 is typically located in a radiology department separate from the imaging bays where the images of the medical imaging examination are acquired.
  • a radiology department may have a number of radiology workstations supporting a staff of radiologists who may work in multiple work shifts.
  • the radiology workstation 18 may also include a server computer or a plurality of server computers, e.g. interconnected to form a server cluster, cloud computing resource, or so forth, to perform more complex image processing or other complex computational tasks.
  • the radiology workstation 18 includes typical components, such as an electronic processor 20 (e.g., a microprocessor, a multi-core processor, a cloud computing resource, or so forth), at least one user input device (e.g., a mouse, touchpad, trackball, or other pointing device 22, a keyboard 23, a microphone 27, and/or the like), and a display device 24 (e.g. an LCD display, plasma display, cathode ray tube display, and/or so forth).
  • the display device 24 can be a separate component from the radiology workstation 18, or may include two or more display devices.
  • the display device 24 of the radiology workstation 18 can comprise a first display device, and a second display device 25 (e.g., an external monitor) can comprise a second display device.
  • a second display device 25 e.g., an external monitor
  • Providing the radiology workstation 18 with two (or more) display devices 24, 25 can be advantageous as it allows one display device to be used to display textual content or other auxiliary information while the other display device is used as a dedicated radiology image viewer; however a radiology workstation with only a single display device is also contemplated.
  • At least one display device 24, 25 of the radiology workstation 18 should be a high-resolution display capable of displaying radiology images I with sufficiently high resolution to enable the radiologist to accurately read the radiology image.
  • first display device 24 and the second display device 25 are used as a dedicated radiology image viewer (namely first display device 24 in the illustrative example), and the other display device (namely second display device 25 in the example) is used to display textual content such as radiology analysis and inputs, and insights from similar past patient cohort analysis and imaging phenotype using radiomics, as well as providing a radiology report entry user interface (UI) 28.
  • UI radiology report entry user interface
  • the electronic processor 20 is operatively connected with one or more non- transitory storage media 26.
  • the non-transitory storage media 26 may, by way of non-limiting illustrative example, include one or more of a magnetic disk, RAID, or other magnetic storage medium; a solid state drive, flash drive, electronically erasable read-only memory (EEROM) or other electronic memory; an optical disk or other optical storage; various combinations thereof; or so forth; and may be for example a network storage, an internal hard drive of the host computer 18, various combinations thereof, or so forth. It is to be understood that any reference to a non- transitory medium or media 26 herein is to be broadly construed as encompassing a single medium or multiple media of the same or different types.
  • the electronic processor 20 may be embodied as a single electronic processor or as two or more electronic processors.
  • the non- transitory storage media 26 stores instructions executable by the at least one electronic processor 20.
  • the instructions include instructions to generate a visualization of at least one graphical user interface (GUI) 28, 29 for display on the display device 24, 25.
  • GUI graphical user interface
  • a radiology GUI 28 can include the radiology report entry user interface 28 and the display window W of an image rendering engine that renders the clinical image(s) I be provided on the radiology display device 24, and a past patient cohort analysis GUI 29 can be provided on the external display device
  • the analytics GUI 29 provides a user input 40, e.g. a search engine input or, more preferably, a chatbot input, and displays information about patients of a cohort retrieved by the data and image analysis, such as an illustrative browsable stack of images 31.
  • a user input 40 e.g. a search engine input or, more preferably, a chatbot input
  • displays information about patients of a cohort retrieved by the data and image analysis such as an illustrative browsable stack of images 31.
  • the radiology reading system 10 is configured as described above to perform a radiology reading method or process 90 and an analogous patient cohort analysis method or process 100.
  • the non-transitory storage medium 26 stores instructions which are readable and executable by the at least one electronic processor 20 of the radiology workstation 18 to perform disclosed operations including performing the radiology reading method or process 90 (for example, rendering the images I in the window W and providing the radiology report entry UI 28) and the analysis method or process 100 accessed via the analytics GUI 29.
  • the methods 90, 100 may be performed at least in part by cloud processing.
  • an illustrative embodiment of the radiology reading method 90 is diagrammatically shown as a flowchart in FIGURE 2.
  • a user e.g., a radiologist, an imaging technologist, surgeon performing an image guidance therapy (IGT) procedure, or other medical personnel logs into the radiology workstation 18 in order to conduct a radiology reading examination or procedure.
  • ICT image guidance therapy
  • a radiology reading environment is provided in which a radiology examination of a patient is opened. This can include retrieving images acquired during the radiology examination from the PACS database 14, and optionally information related to the medical imaging device and/or the radiology procedure from the RIS database 15.
  • an image rendering engine is provided and configured to render the retrieved radiological images of the opened radiology examination on the display device 24, 25, for example as diagrammatically shown in FIGURE 1 by radiology image I rendered in the window W.
  • the radiology report entry UI 28 is provided for receiving a radiology report on the opened radiology examination via the at least one user input device 22, 23, 27 of the radiology workstation 18
  • the report entry UI 28 includes a report window showing at least a portion of the radiology report received from (i.e. entered by) the radiologist on the illustrative display device 25 of the radiology workstation 18.
  • the patient data analysis method 100 is performed by the radiology workstation 18, in which a patient data analytics UI 29 is provided.
  • a patient data analytics UI 29 is provided.
  • the radiology report entry UI 28 is described as being displayed on the display device 24, and the patient data analytics UI 29 is described as being displayed on the external monitor 25 (although the opposite case may be true).
  • the patient data analytics UI 29 should be readily available to the radiologist when the radiologist wants to consult insights from patient data analytics, but otherwise the patient data analytics GUI 29 is preferably unobtrusive.
  • an icon 33 is provided on the radiology workstation which, if selected (e.g., by a mouse click or the like), brings up the patient data analytics UI (e.g. dashboard) 29 with information about the cohort.
  • patient features for the patient of the opened radiology examination are extracted from the EMR database 15.
  • the patient data can include one or more of age, gender, ethnicity, patient symptoms, diagnosis, medications, and laboratory test results, among others. These are merely non-limiting examples, and should not be construed as limiting.
  • examination features for the opened radiology examination are extracted. This can include examination features related to the medical imaging device or modality and/or the radiology procedure from the RIS database 16. Optionally, examination features may include image features automatically extracted from the DICOM metadata associated with the radiology images of the radiology examination being read.
  • the examination features may include radiology findings entered into the radiology report via the UI 28 and detected automatically by natural language processing, keyword detection, or so forth.
  • the examination features may include image features extracted by automated image processing from the radiology images of the radiology examination being read.
  • CAD computer- aided diagnosis
  • a feature may be the absence or presence (and optionally number) of suspicious lesions.
  • a cohort of patients is determined, in which the cohort includes similar patients having patient records stored in the one or more electronic patient databases 14, 15, 16, based on the extracted patient features and the extracted examination features.
  • the operation 106 may be repeated as appropriate to reflect the current set of findings in the report (e.g. as the radiologist adds findings, these may be used to adjust the cohort by repeating operation 106).
  • the patient data analytics UI 29 is provided on the display device 25. In some embodiments, this occurs in response to the radiologist selecting the icon 33. In other embodiments the patient data analytics UI 29 may be displayed in a small window at all times after the cohort is initially determined in operation 106, or the patient data analytics UI 29 may be minimized on a task bar and brought up by user selection of the task bar item, or so forth.
  • the patient data analytics UI 29 includes one or more windows displaying one or more results of the patient data analysis (e.g., the stack of images 31), along with information about the cohort and the user input 40. (In some embodiments the user input may be a microphone 27 receiving user input as speech, in which case the user input may not be shown on a display).
  • the radiology report entry UI 28 and/or the patient data analytics UI 29 can be selectively minimized or expanded on the display device(s) 24, 25 based on inputs received from the user via the at least one user input device 22, 23, 27.
  • FIGURE 3 An example of the patient data analytics UI 29 is shown in FIGURE 3. Further description of the method 100, and in particular with the providing operation 108, will be described with reference to FIGURE 3 (and with continuing reference to FIGURES 1 and 2).
  • the patient data analytics UI 29 can be provided without an input from the user.
  • the patient data analytics UI 29 can automatically be provided on the display device 25.
  • the patient data analytics UI 29 can include a patient data analytics UI dialog 40 for receiving inputs (via the at least one user input device 22, 23, 27) for controlling the performing of the patient data analysis.
  • the patient data analytics UI 29 includes four windows: a window 32 showing data related to different sub-groups (or sub-cohorts) of the cohort of patients (e.g., shown as a pie chart); a window 34 showing data related to patient data of the patient in the cohort (e.g. a cohort patient data window 34- shown as an illustrative bar graph); a window 36 showing data related to radiology results of the patients of the cohort (e.g., a radiology cohort window 36- shown as a bar graph); and window 38 showing images of patients of the cohort (e.g., a cohort image window 38- shown as a grid of thumbnail images; FIGURE 1 shows an alternative way of representing this information as the stack of images 31).
  • a window 32 showing data related to different sub-groups (or sub-cohorts) of the cohort of patients (e.g., shown as a pie chart); a window 34 showing data related to patient data of the patient in the cohort (e.g. a cohort patient data window 34- shown as an illust
  • FIG. 3 The data plots and graphs shown in FIGURE 3 are merely illustrative, and should not be construed as limiting.
  • the user can optionally interact with various of the windows 32, 34, 36, 38 of the patient data analytics UI 29.
  • the user can select (e.g., with a mouse click via the mouse 22, a keystroke with the keyboard 23, or textual instructions with microphone 27, and so forth).
  • the user can select all or a portion of the pie chart of the cohort parameter window 32, or all or a portion of the bar graphs of the cohort patient data window 34 and/or the radiology cohort window 36.
  • an additional field or dialog window can be displayed showing additional information related to the selected portion(s) (e.g., if the “58%” portion of the pie chart in the cohort parameter window 32 is selected, a drop down menu can be provided showing additional details related to the patients of the cohort categorized in this portion of the pie chart).
  • the cohort image window 38 provides a set of thumbnails of radiology images retrieved from radiology examinations of patients of the cohort.
  • the user can select one or more of the thumbnail images, in which a drop-down menu or field or interface dialog may be provided showing additional details related to the patient in the cohort of the selected image.
  • the image rendering engine can be invoked to render an enlarged view of the radiology image corresponding to the thumbnail.
  • an input field or dialog box or window 40 is provided that is selectable by the user to refine parameters of the cohort in the cohort parameter window 32.
  • the user can provide an input via the at least one user input device 22, 23, 27 to enter data into the input to update the cohort parameters.
  • the patient data analytics UI 29 can be updated based on the input data, providing a new pie chart in the cohort parameter window 32 with the updated cohort. With a new cohort, the bar graphs and thumbnail images shown in the windows 34, 36, 38 would be updated accordingly.
  • the input field 40 comprises a freeform text input (e.g., via which the radiologist enters one or more keywords with optional Boolean logic such as “AND” or “OR” connectors).
  • the input field 40 comprises a chatbot configured to receive inputs from the user for controlling the performing of the patient data analysis and/or to update the parameters of the cohort.
  • the user can either enter text via the keyboard 23, select options provided by the chatbot 40 with the mouse 20, or input audio instructions into the chatbot with the microphone 27.
  • the chatbot 40 can provide a series of questions related to the parameters of the cohort on the patient data analytics UI 29. The user can answer these questions with text, mouse clicks, or audio instructions.
  • the chatbot 40 can provide data-driven answers to questions asked by the user.
  • the chatbot 40 can also be used to determine or update the cohort.
  • the chatbot 40 determines a first sub-cohort of patients based on a first medical examination finding (e.g., presence of a brain lesion).
  • a second, different sub-cohort of patients can be determined based on a second, different medical examination finding (e.g., based on gender, and/or age of a patient, and so forth), thus filtering or narrowing the search.
  • the cohort, as displayed in the cohort parameter window can be determined as patients being in both cohorts (e.g., a female over age 50 with a brain lesion).
  • the radiology UI 28 and the patient data analytics UI 29 are merely examples, and should not be construed as limiting.
  • the radiologist is reading a computed tomography (CT) scan of a known liver malignancy.
  • CT computed tomography
  • the radiology system 10 searches and presents the radiologist with a cohort of similar patients, and presents the radiologist with the option to click open the patient data analytics UI 29 that contains the results.
  • the patient response to treatment shows unanticipated atypicality that gives better understanding of the patient’s condition.
  • the radiologist makes a note of it in the report, so the treating physician is aware.
  • the radiologist has a question about prognostic outlook of glioblastoma of 50-year-old Vietnamese female under a specific treatment.
  • the radiologist asks the chatbot 40 the question and enters demographic criteria along with diagnosis.
  • the radiology system 10 accesses the patient data, creates a patient cohort based on the age, gender, ethnicity and diagnosis criteria and presents the results in the patient data analytics UI 29 including images.
  • the imaging phenotype, medications, labs, patient outcomes, response to treatment, etc. for this age, gender and ethnic group cohort are then accessible to the radiologist, who can use these learnings to personalize the prognostication for the current patient.
  • the radiologist has been asked to re-read a case that seems discrepant.
  • the radiologist uses the radiology system 10 to invoke a similar patient/condition cohort to check if retrospectively such a discrepancy had occurred in the past and if it was captured and corrected or was it missed.
  • This retrospective analysis of a re-read driven discrepancy may also point to a pattern of similar discrepancies in the past and perhaps point to a pattern of missed errors and if they have correlation with other clinical patient aspects.
  • the radiologist is diagnosticating a patient with (e.g.
  • the disclosed system 10 allows the user to do the following important functions specifically targeted towards reducing diagnostic errors or discrepancies: changing differential diagnosis, upgrading a diagnosis, increasing suspicion levels for a certain lesion, managing a patient towards better outcomes, among others.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Data Mining & Analysis (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Biomedical Technology (AREA)
  • Databases & Information Systems (AREA)
  • Pathology (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

Un système de lecture de radiologie (10) comprend un poste de travail de radiologie (18) comportant un dispositif d'affichage (24, 25) et au moins un dispositif d'entrée d'utilisateur (22, 23, 27), et au moins un processeur électronique (20) connecté fonctionnellement au dispositif d'affichage et au ou aux dispositifs d'entrée d'utilisateur et programmé de façon : à ouvrir un examen radiologique d'un patient ; à fournir un moteur de rendu d'image configuré afin de rendre des images radiologiques de l'examen radiologique ouvert sur le dispositif d'affichage du poste de travail de radiologie ; et à fournir une interface utilisateur d'entrée de rapport de radiologie (28) pour recevoir un rapport de radiologie sur l'examen radiologique ouvert par l'intermédiaire du ou des dispositifs d'entrée d'utilisateur du poste de travail de radiologie et pour afficher une fenêtre de rapport montrant au moins une partie du rapport de radiologie reçu sur le dispositif d'affichage du poste de travail de radiologie. Le ou les processeurs électroniques sont en outre programmés de façon : à extraire des caractéristiques de patient d'une ou de plusieurs bases de données électroniques de patients (14, 15, 16) pour le patient de l'examen radiologique ouvert ; à extraire des caractéristiques d'examen pour l'examen radiologique ouvert ; à effectuer une analyse de groupe de patients comprenant une analyse radiomique comprenant au moins la détermination d'une cohorte de patients similaires ayant des dossiers de patient stockés dans la ou les bases de données électroniques de patient sur la base des caractéristiques de patient extraites et des caractéristiques d'examen extraites ; à effectuer au moins une analyse de données de patient et une analyse radiomique sur la cohorte ; et simultanément à la fourniture de l'interface utilisateur d'entrée de rapport de radiologie, à fournir en outre une interface utilisateur (UI) d'analyse de données de patient (29) comprenant une fenêtre d'analyse de données de patient affichant au moins un résultat de l'analyse ou des analyses de données de patient sur le dispositif d'affichage du poste de travail de radiologie.
PCT/EP2021/062906 2020-05-20 2021-05-17 Directives de décision de radiologie personnalisées tirées d'une imagerie analogique passée et d'un phénotype clinique applicables au point de lecture Ceased WO2021233795A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US202062704651P 2020-05-20 2020-05-20
US62/704,651 2020-05-20

Publications (1)

Publication Number Publication Date
WO2021233795A1 true WO2021233795A1 (fr) 2021-11-25

Family

ID=76098923

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2021/062906 Ceased WO2021233795A1 (fr) 2020-05-20 2021-05-17 Directives de décision de radiologie personnalisées tirées d'une imagerie analogique passée et d'un phénotype clinique applicables au point de lecture

Country Status (1)

Country Link
WO (1) WO2021233795A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023169812A1 (fr) * 2022-03-08 2023-09-14 Koninklijke Philips N.V. Système de traitement d'imagerie médicale
US12224047B2 (en) 2022-08-31 2025-02-11 Virtual Radiologic Corporation Systems and methods of radiology report processing and display

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120147010A1 (en) * 2010-12-08 2012-06-14 Definiens Ag Graphical User Interface For Interpreting The Results Of Image Analysis
US20130311502A1 (en) * 2011-01-31 2013-11-21 Panasonic Corporation Case searching apparatus and case searching method
US20170053064A1 (en) * 2014-03-03 2017-02-23 Semanticmd, Inc. Personalized content-based patient retrieval system
US20190079938A1 (en) * 2016-03-16 2019-03-14 Koninklijke Philips N.V. Relevance feedback to improve the performance of clustering model that clusters patients with similar profiles together
US20200019617A1 (en) * 2018-07-11 2020-01-16 Google Llc Similar Image Search for Radiology
US20200051699A1 (en) * 2016-10-14 2020-02-13 Koninklijke Philips N.V. System and method to determine relevant prior radiology studies using pacs log files

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120147010A1 (en) * 2010-12-08 2012-06-14 Definiens Ag Graphical User Interface For Interpreting The Results Of Image Analysis
US20130311502A1 (en) * 2011-01-31 2013-11-21 Panasonic Corporation Case searching apparatus and case searching method
US20170053064A1 (en) * 2014-03-03 2017-02-23 Semanticmd, Inc. Personalized content-based patient retrieval system
US20190079938A1 (en) * 2016-03-16 2019-03-14 Koninklijke Philips N.V. Relevance feedback to improve the performance of clustering model that clusters patients with similar profiles together
US20200051699A1 (en) * 2016-10-14 2020-02-13 Koninklijke Philips N.V. System and method to determine relevant prior radiology studies using pacs log files
US20200019617A1 (en) * 2018-07-11 2020-01-16 Google Llc Similar Image Search for Radiology

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LAMBIN P ET AL: "Radiomics: the bridge between medical imaging and personalized medicine", NATURE REVIEWS CLINICAL ONCOLOGY, vol. 14, no. 12, 4 October 2017 (2017-10-04), NY, US, pages 749 - 762, XP055719622, ISSN: 1759-4774, DOI: 10.1038/nrclinonc.2017.141 *
SRIVASTAVA P ET AL: "Automatized Medical Chatbot (Medibot)", 2020 INTERNATIONAL CONFERENCE ON POWER ELECTRONICS & IOT APPLICATIONS IN RENEWABLE ENERGY AND ITS CONTROL (PARC), IEEE, 28 February 2020 (2020-02-28), pages 351 - 354, XP033768706, DOI: 10.1109/PARC49193.2020.236624 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023169812A1 (fr) * 2022-03-08 2023-09-14 Koninklijke Philips N.V. Système de traitement d'imagerie médicale
US12224047B2 (en) 2022-08-31 2025-02-11 Virtual Radiologic Corporation Systems and methods of radiology report processing and display

Similar Documents

Publication Publication Date Title
US11295867B2 (en) Generating and applying subject event timelines
US8335694B2 (en) Gesture-based communication and reporting system
US10949501B2 (en) System and method for compiling medical dossier
JP5317716B2 (ja) 情報処理装置および情報処理方法
US10032236B2 (en) Electronic health record timeline and the human figure
US8655677B2 (en) Productivity workflow index
RU2686627C1 (ru) Автоматическое создание ориентированной на показатели продольной области для просмотра показателей пациента
US10593428B2 (en) Diagnosis support apparatus and method, and non-transitory computer readable medium
US20190156947A1 (en) Automated information collection and evaluation of clinical data
US20100145720A1 (en) Method of extracting real-time structured data and performing data analysis and decision support in medical reporting
JP2023503610A (ja) 医療スキャン映像の相互位置合わせおよびそれとともに使用する方法
US20100131293A1 (en) Interactive multi-axis longitudinal health record systems and methods of use
US20120221347A1 (en) Medical reconciliation, communication, and educational reporting tools
US20100235330A1 (en) Electronic linkage of associated data within the electronic medical record
JP2015524107A (ja) 患者情報を臨床基準にマッチングするシステム及び方法
JP2022528095A (ja) 医療情報を評価するためのプラットフォームおよびその使用方法
US20160267221A1 (en) Medical imaging reference retrieval
US12482553B2 (en) System and method to automatically prepare an attention list for improving radiology workflow
US20180350466A1 (en) Longitudinal health patient profile for incidental findings
US20080175460A1 (en) Pacs portal with automated data mining and software selection
US20130159022A1 (en) Clinical state timeline
JP2019537114A (ja) Ｐａｃｓログ・ファイルを使って関連する過去の放射線医学検査を判別するためのシステムおよび方法
AU2019268083A1 (en) Configuring and displaying a user interface with healthcare studies
US20190198158A1 (en) Automatic Creation of Imaging Story Boards from Medical Imaging Studies
US20230005575A1 (en) Systems and methods for recommending medical tests

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21727788

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21727788

Country of ref document: EP

Kind code of ref document: A1