WO2016138522A1

WO2016138522A1 - Systems and methods for identifying patients at risk for a cardiovascular condition

Info

Publication number: WO2016138522A1
Application number: PCT/US2016/020085
Authority: WO
Inventors: Anthony Carl DEFRANCO; Donald Bryant CONRAD, Jr.
Original assignee: Aurora Health Care Inc
Current assignee: Aurora Health Care Inc
Priority date: 2015-02-27
Filing date: 2016-02-29
Publication date: 2016-09-01
Anticipated expiration: 2017-08-27

Abstract

Systems and methods for identifying patients at risk for a cardiovascular condition are provided. In some aspects, a method includes initiating an image analysis protocol on an image analysis system, and querying, using the image analysis system, a medical record system based a predetermined patient selection criteria. The method also includes receiving from the medical record system data fields consistent with the predetermined patient selection criteria, and retrieving, using the data fields, a plurality of medical images from a medical image storage system. The method further includes analyzing the plurality of medical images to identify patients at risk for a cardiovascular condition, and generating a report indicative of the identified patients.

Description

SYSTEMS AND METHODS FOR IDENTIFYING PATIENTS AT RISK

FOR A CARDIOVASCULAR CONDITION

CROSS-REFERENCE TO RELATED APPLICATIONS

[001] This application claims the benefit of U.S. Provisional Patent Application Serial No. 62/121,947 filed on February 27, 2015, and entitled "SYSTEM AND METHOD FOR CORONARY ARTERY CALCIUM SCORING USING NON-GATED IMAGES."

BACKGROUND

[002] The present disclosure relates generally to medical diagnosis and imaging, and in particular, to systems and methods for imaging and evaluating coronary artery calcification, and other cardiovascular conditions.

[003] About half of cardiovascular deaths in the United States are due to coronary heart disease ("CHD"). Someone dies from a coronary event every minute. Among other risk factors, the direct relationship between the presence and extent of coronary artery calcification and severity of coronary atherosclerosis leading to CHD has long been accepted among medical practitioners. Coronary artery calcium has been detected using various imaging techniques, but in the outpatient setting, it is most commonly evaluated using non-contrast, electrocardiographic ("ECG")- gated, electron-beam computed tomography ("EBCT") or multi-detector computed tomography ("MDCT"). Various studies have demonstrated that EBCT and MDCT can sensitively and reproducibly detect coronary artery calcification.

[004] In general practice, coronary artery calcium is visually identified on ECG-gated images as hyper-attenuating regions with intensities greater than about 130 Hounsfield units extending several contiguous pixels. In order to quantify the calcification burden, a coronary artery calcium score is determined using the Agatston method, whereby each calcified region is assigned a weighted value based on the highest attenuation value, and estimated area of the calcified region. The scores of each region are then summed to give a total score, which is then utilized to determine a grade of coronary artery disease. Generally, absolute Agatston scores less than 10, between 11 and 99, between 100 and 400, and above 400 have been utilized to categorize individuals into groups having minimal, moderate, increased, and extensive amounts of calcification, respectively. Such scoring, has been shown to be highly predictive of future cardiovascular ("CV") events, thus identifying at risk patients in need of treatment.

[005] Conventional, ECG-gated CT scanning requires connecting an ECG lead system to the CT system to coordinate image acquisition during the diastole period of the cardiovascular cycle, when cardiovascular motion is at a minimum, in order to acquire the clearest images for calcium quantification. Such CT scanning is typically performed as part of a risk analysis and treatment assessment for patients with intermediate risk profiles. However, for persons that are asymptomatic, such CT scanning is generally not performed for coronary artery calcium evaluation. Thus, by the time the patient is evaluated for one of the highest predictors of CV events, the patient may already be past the point of prophylactic efforts. Given the rising cost of healthcare and the resulting emphasis on preventative care, this paradigm is ill suited to drastically reducing CHD, despite the known mortality associated with CHD.

[006] Therefore, given that coronary heart disease is the most frequent cause of death in industrialized nations and its onset is currently unpredictable, there is a need for new systems and methods for identifying apparently healthy or asymptomatic individuals at risk for coronary heart disease and other cardiovascular conditions.

SUMMARY

[007] The present disclosure introduces systems and methods for identifying patients at risk for a cardiovascular condition or cardiovascular events using measurements, such as coronary artery calcification, from acquired medical images. In particular, the present disclosure is directed to utilizing medical images that are not especially acquired for the purpose of CV evaluation. As such, systems and methods provided herein may be used to identify cardiovascular risk by estimating coronary artery calcification using non-gated computed tomography ("CT") images. In addition, provided systems and methods may be integrated into various healthcare infrastructures in order to carry out image analysis to identify asymptomatic or non-previously diagnosed patients at risk for cardiovascular events. For instance, medical images stored on a medical image storage system may be periodically, or as prompted, accessed based on predetermined patient selection criteria, and subsequently analyzed.

[008] In accordance with one aspect of the disclosure, a method for identifying patients at risk for a cardiovascular condition is provided. The method includes initiating an image analysis protocol on an image analysis system, and querying, using the image analysis system, a medical record system based on a predetermined patient selection criteria. The method also includes receiving from the medical record system data fields consistent with the predetermined patient selection criteria, and retrieving, using the data fields, a plurality of medical images from a medical image storage system. The method further includes analyzing the plurality of medical images to identify patients at risk for a cardiovascular condition, and generating a report indicative of the identified patients.

[009] In accordance with another aspect of the disclosure, a method for generating a report of a patient's cardiovascular risk using coronary calcification estimated from non-gated computed tomography ("CT") images using an image analysis system is provided. The method includes accessing, from a medical image storage system, a set of non-gated CT images acquired from a patient using a CT system, and identifying at least one calcified region using the set of non-gated CT images. The method also includes estimating a calcification score using the at least one calcified region identified, and determining the patient's cardiovascular risk using the estimated calcification score. The method further includes generating a report indicative of the patient's cardiovascular risk. [0010] In accordance with another aspect of the disclosure, a system for identifying patients at risk for a cardiovascular condition is provided. The system includes a medical record system having stored therein patient records, and a medical image storage system having stored therein medical images. The system also includes an image analysis system configured to carry out an image analysis protocol by querying the medical record system based a predetermined patient selection criteria, and receiving from the medical record system data fields consistent with the predetermined patient selection criteria. The image analysis system is also configured to carry out steps for retrieving, using the data fields, a plurality of medical images from a medical image storage system, and analyzing the plurality of medical images to identify patients at risk for a cardiovascular condition. The image analysis system is further configured to carry out steps for generating a report indicative of the identified patients.

[0011] The foregoing and other advantages of the invention will appear from the following description.

BRIEF DESCRIPTION OF THE DRAWINGS

[0012] FIG. 1 shows a typical imaging study workflow typically carried out in a healthcare institution.

[0013] FIG. 2 is a diagram showing an example system, in accordance with aspects of the present disclosure.

[0014] FIG. 3 is a flowchart setting forth steps of a process in accordance with aspects of the present disclosure.

[0015] FIG. 4 shows steps of another process in accordance with aspects of the present disclosure.

[0016] FIG. 5 is another flowchart setting forth steps of a process, in accordance with aspects of the present disclosure.

[0017] FIG. 6A and 6B illustrate an example CT system, in accordance with aspects of the present disclosure. [0018] FIG. 7A is a graph showing a correlation between calcification scores estimated using gated and non-gated CT images.

[0019] FIG. 7B is another graph showing a correlation between calcification scores estimated using gated and non-gated CT images.

DETAILED DESCRIPTION

[0020] Turning to FIG. 1, the general steps of a process 100 associated with an imaging study workflow typically carried out in a healthcare institution are illustrated. The process 100 usually begins with a physician or medical professional ordering an imaging study to be performed on a patient, as indicated by process block 102. Example imaging studies can include computed tomography ("CT") imaging, magnetic resonance ("MR") imaging, positron emission tomography ("PET") imaging, ultrasound ("US)" imaging, and others.

[0021] As a result, an order record is created in an Electronic Medical Record ("EMR") system, as indicated by process block 104. The record may include information regarding the imaging study being ordered, as well as information about the patient and ordering provider. The record may also include details about the type of image study to be performed, such as preferences for how the imaging data is to be processed or reconstructed. For example, an ordering physician or medical professional might want to view an anatomical area of interest from multiple points of view (e.g. axial, sagittal, and coronal).

[0022] Then, the imaging study is executed on one or more imaging systems, in accordance with specifications stored in the record, as indicated by process block 106. For example, an order to produce images of a patient's thoracic region for medical analysis may be carried out using a CT system, for example, as described with reference to FIGs. 6A and 6B. Depending upon specifications in the order record, the study's acquired volumetric data may be reconstructed and split into separate "series." For instance, a single study may yield multiple series, each representing different anatomical views, or time sequences. As such, an imaging study might include up to 1,000 individual images, or more. The reconstructed images, grouped by series, of the imaging study can then be provided to a workstation for analysis by a medical professional, as indicated by process block 108.

[0023] In some healthcare institutions, a separate technology solution may be utilized to store images associated with a particular study, as indicated by process block 110. The storage of the images in block 110 may occur in tandem with, in conjunction with, or after the analysis in block 108. For example, a medical professional may perform an analysis on the images as indicated in block 108 by using a workstation that is directly connected to a CT system while a copy of the images and series is independently stored on a medical image storage system. In another example, a system may be configured to send the image information to the medical image storage system first and the analysis outlined in block 108 could be performed using a workstation that is linked via network connection to the medical image storage system. In situations where this additional storage exists, images, along with other study information, may be stored on a server or centralized data repository, to be later retrieved for analysis.

[0024] Then, at process block 112, a physician, radiologist or other medical professional upon review the imaging study based on the order specifications, would then report findings by adding the results into the medical record stored on the EMR system. Often, the order record in the EMR also contains discrete data and text fields that allow the results and interpretation of the study to be part of the same order record. Importantly, EMR systems can often be configured so that study order records contain fields for results that may or may not directly pertain to the medical question that precipitated the original order. For example, a study of a patient's thoracic region using ungated CT images may be ordered to analyze a pulmonary defect or for lung cancer screening. The order record would contain the results of the analysis for that diagnostic purpose, but may also contain fields for the results of separate analyses, such as analyzes conducted using study images stored on a medical image storage system. This configuration has important implications for the use of existing images to detect non-previously diagnosed conditions, for example, in patients that may be at risk for coronary heart disease ("CHD").

[0025] Various studies have focused on risk assessment in symptomatic and asymptomatic populations, including the role of coronary artery calcium ("CAC") scoring in assessment of symptomatic patients. These have typically utilized coronary CT images to evaluate the extent of coronary atherosclerosis, as well as determine the cost-effectiveness of coronary calcium scoring for assessment of cardiovascular death or myocardial infarction. In particular, quantifying coronary calcification has involved acquiring and evaluating ECG-gated CT images, a procedure mainly applied to symptomatic patients or those considered at intermediate risk for cardiovascular disease.

[0026] However, more than 10 million non-gated chest CT scans are performed in the United States each year to diagnose and treat various non-cardiovascular conditions. For example, non-gated CT images may be acquired during a lung cancer screening. Importantly, although some indicators of coronary calcification may be present in such non-gated CT images, at best, these might be "incidentally noted" on radiology reports, or worse, not mentioned at all. Also, the extent of incidentally discovered coronary artery calcification would not be clinically quantified.

[0027] Presently, use of non-gated CT images for determining coronary calcification is not widely implemented because the research that supports the CAC as a prognostic risk indicator is almost exclusively based on ECG-gated CT studies. Hence, standard clinical practice utilizes ECG-gated CT scans for the determination of CAC burden. In addition, the lack of clinically validated data that correlate non- gated CAC scores with gated CAC scores, as will be described with reference to FIGs. 7A and 7B, has presented further challenges in implementing non-gated CT images for determining coronary calcification. Hence, non-gated scans, while more prolific as a diagnostic modality, have not be considered as a viable method for CAC scoring and predicting cardiovascular risk. [0028] Therefore, in some aspects of the disclosure, it is recognized that imaging typically used for diagnosing non-cardiovascular conditions, such as non-gated CT images, could be advantageously analyzed to determine a patient's cardiovascular risk. Therefore, the present disclosure provides systems and methods directed to using non-gated CT images to predict the risk of coronary and/or vascular events using measures of coronary artery calcification. As detailed in the example below, this approach provides results that are clinically comparable to those obtained using a dedicated, ECG-gated CT scanner.

[0029] Specifically, a study of a large population was performed to determine the accuracy of determining calcification burden and cardiovascular risk using non- gated CT images. Using the AHA 17-segment coronary artery classification system, and a threshold of 130 HU for calcified plaque, each visible plaque was manually circled on both gated and non-gated images. For each participant, a workstation was used to automatically compute a CAC score for each coronary vessel, and a total CAC score was then computed for the entire cardiovascular scan. CAC scores from a random subset of the large population were then grouped according to Agatson CAC units into four quartiles, namely 0, 1-100, 101-400, and greater than 400, with 135 participants having CAC of roughly zero, and 381 participants having scores greater than 0. As shown in FIGs. 7A and 7B, a linear regression demonstrated a strong correlation between results obtained using ECG-gated cardiovascular images and non-gated images.

[0030] In addition, results of the study showed that with non-gated scanning 439 participants (85%) were correctly classified into their CAC quartile, while only 77 (15%) participants were reclassified. Specifically, mild CAC scores were misread in 2% of non-gated scans. Over-estimation of CAC on non-gated scans were uncommon and reclassified only one quartile higher, which is unlikely to change risk stratification clinically. With CAC scores over 100, misclassification was uncommon but more likely to be due to under-reading rather than over-reading, which may alter risk stratification. CAC scores over 400 were rarely interpreted as mild (0.4%) and intermediate calcification (1.2%). These results demonstrate that chest CT scans, and other scans, may provide an opportunity to identify patients at higher risk for subsequent CV events, thus allowing for initiation of targeted primary prevention measures.

[0031] Thus, the present disclosure not only extends the clinical application of non-gated CT imaging, but also unifies disparate applications of underutilized technology. In some aspects, the provided systems and methods overcome the need for dedicated, ECG-gated cardiovascular CT scans, as conventionally utilized for estimating the presence and extent of coronary atherosclerosis, and, hence, risk of myocardial infarction ("MI"), CV death, and other CV events. This approach potentially avoids the additional radiation exposure, cost, and time associated with obtaining coronary calcification information by traditional ECG-gated CT scanning, by making use of images already acquired. In addition, patients with previously unknown coronary atherosclerosis, who have a substantially higher risk of MI and CV complications in the short and long term, may also be diagnosed. Furthermore, future health care costs for apparently healthy or asymptomatic individuals may be substantially reduced by identifying and modifying risk factors in individuals with previously undiagnosed disease.

[0032] In other aspects of the disclosure, systems and methods provided herein may be used to selectively analyze a patient population having received various medical images, in order to identify patients suffering from certain cardiovascular conditions or at risk from cardiovascular events. More particularly, as will be described, medical images, such as non-gated CT images stored on a medical image storage system, may be periodically, or as prompted, accessed according to a predetermined patient selection criteria for analysis. As such, at risk patients for cardiovascular events may be identified based on estimated calcification.

[0033] Referring to FIG. 2, a system 200, in accordance with aspects of the present disclosure, is shown. In general, the system 200 may include a medical record system 202, an image analysis system 204, and a medical image storage system 206, all connected using one or more wired or wireless connections allowing communication and transfer a variety of data, images, and other information therebetween, as shown. In some implementations, the image analysis system 204 may be a laptop, tablet, person computer, workstation, server, or other computing system or device. In this sense, the image analysis system 204 can have a broad range of functionality and hardware, and be configured to or include non-transient instructions for carrying out steps in accordance with aspects of the disclosure.

[0034] The medical record system 202 may include an operational system 208, a reporting system 210 and a data storage 212, such as a data warehouse or data registry, as shown in FIG. 2. As described, the medical record system 202 may include a variety of information and data, including order records, patient information, provider information, study information, and so forth. The medical record system 202 may be configured to allow transfer of such information and data between the operational system 208, reporting system 210 and data storage 212, as needed.

[0035] In some implementations, the image analysis system 204 may include a routing module 214, an analysis module 216 and a storage module 218. Specifically, the routing module 214, as directed by the analysis module 216, is configured to communicate with the medical record system 202 and medical image storage system 206 to selectively retrieve medical images for analysis, in accordance with a predefined patient selection criteria. In some aspects, non-gated CT images may be retrieved from the medical image storage system 206 and analyzed for calcification using the image analysis module 216. In this manner, patients at risk for cardiovascular conditions or events may be identified. By way of example, the medical image storage system 206 may include a PACS or Enterprise storage system. It may be appreciated that although the routing module 214, analysis module 216 and storage module 218 forming the image analysis system 204 are shown in FIG. 2 as separate components, in some implementations, these may be combined into a single system.

[0036] As shown in FIG. 2, in some aspects, images retrieved from the medical image storage system 206 may be transferred or copied to an intermediary storage module 218, for temporary storage for the duration of analysis, for instance. Other data and information may also be stored in the storage module 218, including work queues, analysis results, and so forth. Such information, may be relayed back to the medical record system 202 via the routing module 214 and the direction of the analysis module 216. For instance, analysis results regarding a patient's calcification burden or score, or a risk for a cardiovascular condition or event may be included into the patient record in the medical record system 202. In some aspects, alerts or follow-up instructions for a primary physician or clinician may also be included in the patient record based on results from an image analysis protocol, in accordance with the present disclosure.

[0037] Referring now to FIG. 3, steps of a process 300 in accordance with aspects of the present disclosure are illustrated. The process 300 may carried out using a system as described with reference to FIG. 2. Specifically, the process 300 may begin at process block 302 with initiating an image analysis protocol to identify patients at risk for certain cardiovascular conditions or events. In some aspects, the image analysis protocol may be carried out periodically, or aperiodically. In other aspects, the image analysis protocol may be carried out following a user prompt or in response to an event, such as receiving a signal indicating an update to a database or medical image storage system having been performed. As appreciated from above, the image analysis protocol may be carried out in an automated or semi-automated fashion.

[0038] At process block 304, a query may be carried out on a medical record system using a predetermined patient selection criteria. Referring particularly to FIG. 2, any or all of the operational system 208, the reporting system 210, and data storage 212 may be queried. The query can be in the form of electronic signals or instructions from the image analysis system 204 to a medical record system 202. In particular, the query may be configured to select patients that match the predetermined patient selection for analysis. For instance, the query may selectively filter patients that have had certain imaging studies, such as non-gated CT images. In some aspects, a record of the query might also be generated in a patient record at process block 304. [0039] Also, a history of selected surgical procedures or device implantation may be included in the query, such as previous open heart surgery or previous interventional cardiology measures. This would be advantageous because patients with previous surgeries are more likely to already have known issues and may have metal implants that could skew and distort x-ray energy, thereby decreasing the reliability of the automated calcification score. A prior history of encounters with cardiovascular physicians might also be included in a query. Specifically patients with high levels of CAC that have no known previous history of visits with cardiovascular physicians would be more likely to be undiagnosed. Other filters in the query may include self-reported symptoms, such as reported chest pain, referrals, diagnoses. In addition, results from previous analyses may also be included in the query. For example, imaging studies where a CAC score has already been obtained might be excluded. Similarly, studies for patients previously screened by an image analysis protocol might also be excluded.

[0040] Furthermore, opt-in or opt-out filters may also be included in the query carried out at process block 304. For instance, in some implementations, a clinician might receive a notification or prompt to opt-in or opt-out of an automated image analysis protocol, in accordance with aspects the present disclosure, when entering an imaging order for a patient. This would allow for situation where, for example, a patient may not consent to using images for subsequent analysis. As such, data from such patient might be included or excluded at process block 304 based on the clinician selection.

[0041] Following the query, a plurality of data fields, consistent with the predetermined patient selection criteria, may then be received from the medical record system by the image analysis system 204, as indicated by process block 306. The data fields may include patient information or demographics, such as a medical record ID, age, gender, and so forth, as well as imaging study-specific identifiers, such as accession number, medical records system study ID, order ID, study date, and so forth. The data fields may also include consent or user opt-in or opt-out flags. Information associated with the plurality of data fields may be provided to the image analysis system 204, and more particularly the routing module 214, through any number of known communication methods. Examples of such methods include HL7, SOAP, REST, SQL, and others. Information can also be imported from standalone files exported from the medical record system 202 in any number of formats such as XML, JSON, CSV, TSV, and others.

[0042] At process block 308, a plurality of medical images, potentially associated with one or more patients, may then be retrieved using the received data fields by the analysis system 204, as described with reference to FIG. 2. In particular, the routing module 214 may be configured to utilize the data fields received at process block 308 to query the medical image storage system 206. In some aspects, the medical image storage system 206 may be queried on several levels, including patient, study and series. This process is illustrated in FIG. 4.

[0043] In the first step 402 shown in FIG. 4, the medical image storage system 206 is interrogated by the routing module 214 using patient and study information in order to translate the medical record information or data fields into a corresponding Root DICOM or equivalent identifier. This allows images associated with a desired study to be located in the medical image storage system 206. In some aspects, step 402 can be accomplished using a C-Find function directed to DICOM formats or an equivalent for XDS, or other standardized medical imaging formats. The routing module 214 uses the results from the first query to perform an additional filtering step to confirm that the study imaging characteristics, such as gantry speed, slice thickness, x-ray amperage, image file format, and so forth, fit desired specifications. If no error is returned, the routing module 214 can then interrogate the medical image storage system 206 a second time, as indicated by step 404.

[0044] In the second query, based on a study root ID from the first query, series information can then be obtained. Similarly, a C-FIND function or equivalent may be utilized in the second query. For example, if a machine learning algorithm trained to estimate coronary calcification on axial-oriented non-gated CT images, the routing module 214 would then be configured to use C-Find or equivalent query method to determine the unique DICOM or equivalent series identifier corresponding to an axial reconstruction of the raw volumetric data among the all of the images represented by the study root ID. In scenarios when no eligible images are found, the process would be stopped, and an absence of an eligible study would be recorded, for instance, in the storage module 218, described with reference to FIG. 2, for example, in the form of a negative eligibility flag. The next eligible study in the work queue would then be pursued.

[0045] Referring again to step 308 of FIG. 3, the routing module 214 would then take the results of the second query and initiate a copy procedure to transfer the desired images or image series to the storage module 218, to be accessed by the analysis module 216. In some implementations, it may be possible to embed scanning directly into medical image storage system 206, such as a PACS or Enterprise Medical Image Storage System. However, given the variability of how databases in such systems are constructed, and fact that most image systems are generally FDA approved, and therefore non modifiable, it may be more convenient to copy the desired series of images for analysis. This provides a more versatile approach, since open standards such as DICOM and XDS make querying a variety of PACS / Enterprise Medical Image Storage Systems more feasible. To initiate the copy, the routing module 214 can initiate the C-MOVE DICOM function or XDS equivalent, directing the copied images or image series to the storage module 218. In addition to storing a copy of the desired images, the storage module 218 can also persist the query results.

[0046] Referring again to FIG. 3, the medical images retrieved from the storage module 218 can then be analyzed by the analysis module 216 to identify patients at risk for a cardiovascular condition, as indicated by process block 310. In some aspects, a risk for a cardiovascular condition or event may be determined at process block 310. For instance, one or more calcification scores may be estimated using non-gated CT images in order to determine a patient's cardiovascular risk, as will be described. In some aspects, the storage module 218 may retain the copied image data following the analysis. Alternatively, to minimize storage space requirements, the storage module 218 may delete the copied image data, while retaining the analysis results and generating timestamp information for auditing and reporting purposes.

[0047] Then, at process block 312, a report, of any form, may then be generated indicating results of the analysis. For example, the report may be in the form of a list or include information about patients at risk for certain cardiovascular conditions, or in need for specific interventions. Referring particularly to FIG. 2, the routing module 214 may receive the report generated by the analysis module 216. Specifically, the routing module 214 may take pre-specified elements of the report and re-package them in messaging formats compliant with communication protocols such HL7, SOAP, REST, SQL and so forth, to be transmitted to the medical record system 202. Results can also be exported into standalone files in any number of formats including as XML, JSON, CSV, TSV, or others that can be read into the medical record system 202 and appended to the appropriate patient records. The results can then become part of patient records, thereby allowing subsequent viewing by a clinician and creating the ability to include the results in other reports and work queues derived from the operational system 208, the reporting system 208, or the data storage system 212.

[0048] In some aspects, the report can be sent by the routing module 214 to pre- specified recipients, for instance via a pre-formatted message sent securely via email or other electronic communication method. The report may also be provided via an RPE or other interface format to create a work queue or execute a remote process within a records management system to notify a user or physician of test results. In some aspects, the report may be utilized in a further analysis, to generate correlations between certain patient characteristics and identified conditions.

[0049] Turning now to FIG. 5, the steps of a process 500 in accordance with aspects of the present disclosure are illustrated. The process 500 may be performed on any of a variety of suitable systems or processing devices, such as system 200 described with reference to FIG. 2. [0050] The process 500 may begin at process block 502 by receiving a plurality of medical images, such as non-gated CT images, associated with one or more patients. In some aspects, volumetric non-gated CT image data may be acquired and reconstructed at process block 502, for example, using a CT system, as will be described. On the other hand, the medical images may be previously-reconstructed images, and accessed at process block 502, for example, from a medical image storage system, database or other storage medium, as described. In some aspects, acquired or accessed images may also be pre-processed at process block 502. For instance, images may be pre-processed using a variety of digital contrast enhancement techniques to form an index or set of weighting factors for determining the likelihood that calcification is present within a specific region of the image.

[0051] At process block 504, accessed or acquired images may then be analyzed. In particular, non-gated CT images may be analyzed to identify regions, or volumes, of calcification. In analyzing the images, a boundary estimation may be performed on non-gated CT images, for example, using an algorithm derived from machine learning or computer vision techniques. Other segmentation algorithms, threshholding algorithms, and other methods for identifying regions of calcification may also be utilized, which may or may not require manual input from a clinician. In some aspects, a logical filter may be utilized at process block 504 to control against false identification of calcified regions. For example, a user or operator may be prompted to provide information regarding artifacts generated by implants or devices, such as a stent.

[0052] In addition, a feature orientation and a scale determination may be performed at process block 504. As such, a variety of information related to the calcified regions identified, including highest attenuation value, area, volume, and so forth, may be obtained and/or analyzed at process block 504. Although process blocks 502 and 504 describe use of non-gated CT images, it may be envisioned, however, that any combination of gated and/or non-gated images, may be utilized at process blocks 502 and 502 to identify calcified regions. In addition, such images need not be directed to depictions of coronary structures. For instance, as described, images of non- symptomatic patients, obtained from non-coronary studies, may be advantageously utilized in process 500.

[0053] Then, at process block 506, a calcification score may be estimated for each of the calcified regions using the calcification information obtained from process block 504. For example, the calcification score may include an Agatston score, a calcium volume score, a calcium mass score, and so forth. In some aspects, process block 506 may include assigning a density factor based upon the highest pixel or voxel intensity found in the identified calcified regions or volumes, and multiplying the assigned density factor by the area or volume of the calcification to arrive at the calcification score. A total calcification score may also be computed by summing the calcification scores for each individual calcified region.

[0054] At process block 508, a patient's cardiovascular risk may then be determined using the estimated calcification scores. Such cardiovascular risk can include a risk for at least one of a myocardial infarction, or a cardiovascular death, as non-limiting examples. In some aspects, a user or operator may be prompted to provide information related to the patient, including the patient's age, gender, race, as well as other factors affecting the patient's cardiovascular risk. Additionally, or alternatively, such information may be obtained from a medical record system, or other system, storage, or database, as described. In this manner, an age-adjusted quartile risk may be determined at process block 508 using the estimated calcification scores.

[0055] Finally, at process block 510 a report may be generated. The report may take a variety of forms. In particular, the report may indicate the patient's cardiovascular risk. The report may also provide illustrations indicative of identified calcified regions as well as estimated calcification scores. In some aspects, an indication may be provided to a user or operator when a calcification score exceeds a threshold value. For example, an indication regarding outliers with scores greater than about 400 may be provided to identify high-risk patients. In other aspects, the report related to a cardiovascular risk, or calcification score may be delivered electronically to a medical record system or a clinician, or other suitable system, as described.

[0056] As appreciated form the above, process 500 may be automated and/or integrated with a variety of computer or hardware systems. For example, the above-described process 500 may be configured to be performed automatically by a computer system that is networked within a healthcare institution and configured to analyze coronary CT images. As described, process 500 may be carried out periodically, or aperiodically, or as a result of a user prompt or in response to an event, such as signal indicating an updated database or medical image storage system. Additionally or alternatively, the above-described process 500 may be configured to be performed automatically by a computer system that is networked with or integrated as part of an imaging or image analysis system or workstation.

[0057] Referring particularly now to FIGS. 6A and 6B, an example of an X-ray computed tomography ("CT") imaging system 600, for use in accordance with aspects of the present disclosure, is illustrated. The CT system 600 includes a gantry 602 that extends about a bore 604 and includes at least one X-ray source 606. The CT system 600 shown in FIG. 6A also includes detector modules 608 arranged diametrically opposed to the X-ray source 606, which comprise multiple X- ray detector elements 610. In some CT system implementations (not shown in FIGS. 6A and 6B), the detector modules 608 are circumferentially arranged all around the gantry 602.

[0058] During operation, the X-ray source 606 projects an X-ray beam 612, which may be a fan-beam or cone-beam of X-rays, towards respective detector modules 608. Together, the X-ray detector elements 610 sense the projected X-rays 612 that pass through or are scattered by a patient 614, such as a medical patient or an object undergoing examination, which is positioned in the bore 604 of the CT system 600. Each X-ray detector element 610 produces an electrical signal that may represent the intensity of an impinging X-ray beam and, hence, the attenuation or scattered components of the beam as they passes through or scatter from the patient 614. In some configurations, each X-ray detector 610 may be capable of counting the number of X-ray photons that impinge upon the detector 610. That is, the detector 610 may include photon counting and/or energy discriminating detectors. However, the detector 610 may also include energy integrating detectors.

[0059] The CT system 600 also includes an operator workstation 616, which typically includes a display 618; one or more input devices 620, such as a keyboard and mouse; and a computer processor 622. The computer processor 622 may include a commercially available programmable machine running a commercially available operating system. The operator workstation 616 provides the operator interface that enables scanning control parameters to be entered into the CT system 600. The operator workstation 616 may be in communication with a data store server 624 and an image reconstruction system 626, or the functions of data storage and image reconstruction may be performed on the operator workstation 616. By way of example, the operator workstation 616, data store sever 624, and image reconstruction system 626 may be connected via a communication system 628, which may include any suitable network connection, whether wired, wireless, or a combination of both. As an example, the communication system 628 may include both proprietary or dedicated networks, as well as open networks, such as the internet.

[0060] The operator workstation 616 is also in communication with a control system 130 that controls operation of the CT system 600. The control system 630 generally includes an X-ray controller 632 and, optionally, may include a table controller 634, an optional gantry controller 636, and a data acquisition system 638. The X-ray controller 632 provides power and timing signals to the X-ray source 606 to achieve a desired X-ray exposure. If included, the table controller 634 controls a table 640 to position the patient 612 in the gantry 602 of the CT system 600. In some situations, the table controller 634 may be mechanically controlled, such as by manually actuated levers or other controls. Furthermore, powered or manual control may be foregone. For example, in some situations, the table 640 may be a stretcher or make-shift table and the patient and table 640 adjusted as needed, such as may occur in battlefield or other deployments. In some alternative configurations, an optional gantry controller 636 may control the position of the gantry 602 with respect to the patient 614.

[0061] The data acquisition system ("DAS") 638 samples data from the detector modules 608 and converts the data to digital signals for subsequent processing. In some aspects, data acquisition is gated, or timed to a cardiovascular cycle, or periods therein, for example, using an electrocardiogram ("ECG") signal. In other aspects, data acquisition is not gated. Digitized X-ray data may then be communicated from the DAS 638 to the data store server 624. By way of example, the data store server 624, may include a local storage server, or a storage module and/or a medical image storage system, as described with reference to FIG. 2.

[0062] The image reconstruction system 626 then retrieves the X-ray data from the data store server 624 and reconstructs an image therefrom. The image reconstruction system 626 may include a commercially available computer processor, or may be a highly-parallel computer architecture, such as a system that includes multiple-core processors and massively parallel, high-density computing devices. Optionally, as mentioned, image reconstruction can also be performed on the processor 622 in the operator workstation 616. If not reconstructed at the operator workstation 616, reconstructed images can be communicated back to the data store server 624 for storage or to the operator workstation 616 to be displayed to the operator or clinician.

[0063] The CT system 600 may also include one or more networked workstations 642. By way of example, a networked workstation 642 may include a display 644; one or more input devices 646, such as a keyboard and mouse; and a processor 648. The networked workstation 642 may be located within the same facility as the operator workstation 616, or in a different facility, such as a different healthcare institution or clinic. In some aspects, the networked workstations 642 may be configured to receive and process a set of reconstructed CT images acquired from the patient 614. In addition, in some aspects, the networked workstations 642 may also be configured or programmed to identify and characterize calcified regions in the CT images, and estimate calcification scores using the calcified regions identified. The networked workstations 642 may also be configured or programmed to determine and report a cardiovascular risk of the patient 614 by using estimated calcification scores. In some alternative implementations, the operator workstation 616 may be configured to perform any of the above steps, including alerting or reporting to a clinician the presence and/or extent of coronary calcification, as well as the cardiovascular risk of the patient. For instance, a clinician may be alerted when a high-risk patient is identified.

[0064] The networked workstation 642, whether within the same facility or in a different facility as the operator workstation 616, may gain remote access to the data store server 624 and/or the image reconstruction system 626 via the communication system 128. Accordingly, multiple networked workstations 642 may have access to the data store server 624 and/or image reconstruction system 626. In this manner, x-ray data, reconstructed images, or other data may be exchanged between the data store server 624, the image reconstruction system 626, and the networked workstations 642, such that the data or images may be remotely processed by a networked workstation 642. This data may be exchanged in any suitable format, such as in accordance with the transmission control protocol ("TCP"), the internet protocol ("IP"), or other known or suitable protocols.

[0065] Features suitable for such combinations and sub -combinations would be readily apparent to persons skilled in the art upon review of the present application as a whole. The patient matter described herein and in the recited claims intends to cover and embrace all suitable changes in technology.

Claims

1. A method for identifying patients at risk for a cardiovascular condition, the method comprising:

initiating an image analysis protocol on an image analysis system;

querying, using the image analysis system, a medical record system based on a predetermined patient selection criteria;

receiving from the medical record system data fields consistent with the predetermined patient selection criteria;

retrieving, using the data fields, a plurahty of medical images from a medical image storage system;

analyzing, using the image analysis system, the plurality of medical images to identify patients at risk for a cardiovascular condition; and

generating, using the image analysis system, a report indicative of the identified patients based on the analyzing the plurality of medical images to identify patients at risk for the cardiovascular condition.

2. The method of claim 1, wherein the image analysis protocol is automatically initiated periodically or aperiodic ally.

3. The method of claim 1, wherein the image analysis protocol is initiated following a user prompt or in response to an event.

4. The method of claim 1, wherein the plurality of medical images comprises non-gated computed tomography ("CT") images.

5. The method of claim 4, wherein analyzing the plurality of medical images comprises identifying at least one calcified region using the non-gated CT images and estimating a calcification score using the at least one calcified region identified.

6. The method of claim 5, wherein the calcification score includes an Agatston score, a volume score, or a calcium mass score, or a combination thereof.

7. The method of claim 5, wherein the method further comprises determining a patient's cardiovascular risk using the calcification score.

8. The method of claim 1, wherein the method further comprises updating the medical record system using information in the report.

9. The method of claim 1, wherein the method further comprises providing an indication alerting a clinician of the cardiovascular condition for the identified patients.

10. A method for generating a report of a patient's cardiovascular risk using coronary calcification estimated from non-gated computed tomography ("CT") images using an image analysis system, the method comprising:

i) accessing, from a medical image storage system, a set of non-gated CT images acquired from a patient using a CT system;

ii) identifying at least one calcified region using the set of non-gated CT images;

iii) estimating a calcification score using the at least one calcified region identified;

iv) determining the patient's cardiovascular risk using the estimated calcification score; and

v) generating a report indicative of the patient's cardiovascular risk.

11. The method of claim 10, the method further comprising processing the non-gated CT images by performing at least one of a contrast enhancement or an indexing algorithm.

12. The method of claim 10, wherein step ii) further includes performing a feature orientation and a scale determination.

13. The method of claim 10, wherein step ii) further includes performing a calcification boundary estimation using a machine learning or computer vision algorithm.

14. The method of claim 10, wherein the calcification score includes an Agatston score, a volume score, or a calcium mass score, or a combination thereof.

15. The method of claim 10, wherein the patient's cardiovascular risk includes a quartile risk.

16. The method of claim 10, the method further comprising receiving an input related to the patient's age.

17. The method of claim 16, the method further comprising determining an age-adjusted quartile risk using the input.

18. The method of claim 10, wherein the patient's cardiovascular risk includes a risk for at least one of a myocardial infarction, or a cardiovascular death, or both.

19. The method of claim 10, the method further comprising providing an indication when the estimated calcification score of the patient is greater than about 400.

20. A system for identifying patients at risk for a cardiovascular condition, the system comprising:

a medical record system having stored therein patient records;

a medical image storage system having stored therein medical images; and an image analysis system configured to carry out an image analysis protocol by:

querying the medical record system based a predetermined patient selection criteria;

retrieving, using the data fields, a plurality of medical images from a medical image storage system;

analyzing the plurality of medical images to identify patients at risk for a cardiovascular condition; and

generating a report indicative of the identified patients.

21. The system of claim 20, wherein the image analysis protocol is initiated periodically or aperiodically.

22. The system of claim 20, wherein the image analysis protocol is initiated following a user prompt or in response to an event.

23. The system of claim 20, wherein the plurality of medical images comprises non-gated computed tomography ("CT") images.

24. The system of claim 23, wherein the image analysis system is further configured to identity at least one calcified region using the non-gated CT images and estimate a calcification score using the at least one calcified region identified.

25. The system of claim 24, wherein the calcification score includes an Agatston score, a volume score, or a calcium mass score, or a combination thereof.

26. The system of claim 25, wherein the image analysis system is further configured to determine a cardiovascular risk using the calcification score.

27. The system of claim 20, wherein the image analysis system is further configured to update the medical record system using information in the report.

28. The system of claim 20, wherein the image analysis system is further configured to provide an indication alerting a clinician of the cardiovascular condition for the identified patients.