Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography
Abstract Chest computed tomography is one of the most common diagnostic tests, with 15 million scans performed annually in the United States. Coronary calcium can be visualized on these scans, but other measures of cardiac risk such as atrial and ventricular volumes have classically required adminis...
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-03-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-46977-3 |
| _version_ | 1797233454392279040 |
|---|---|
| author | Robert J. H. Miller Aditya Killekar Aakash Shanbhag Bryan Bednarski Anna M. Michalowska Terrence D. Ruddy Andrew J. Einstein David E. Newby Mark Lemley Konrad Pieszko Serge D. Van Kriekinge Paul B. Kavanagh Joanna X. Liang Cathleen Huang Damini Dey Daniel S. Berman Piotr J. Slomka |
| author_facet | Robert J. H. Miller Aditya Killekar Aakash Shanbhag Bryan Bednarski Anna M. Michalowska Terrence D. Ruddy Andrew J. Einstein David E. Newby Mark Lemley Konrad Pieszko Serge D. Van Kriekinge Paul B. Kavanagh Joanna X. Liang Cathleen Huang Damini Dey Daniel S. Berman Piotr J. Slomka |
| author_sort | Robert J. H. Miller |
| collection | DOAJ |
| description | Abstract Chest computed tomography is one of the most common diagnostic tests, with 15 million scans performed annually in the United States. Coronary calcium can be visualized on these scans, but other measures of cardiac risk such as atrial and ventricular volumes have classically required administration of contrast. Here we show that a fully automated pipeline, incorporating two artificial intelligence models, automatically quantifies coronary calcium, left atrial volume, left ventricular mass, and other cardiac chamber volumes in 29,687 patients from three cohorts. The model processes chamber volumes and coronary artery calcium with an end-to-end time of ~18 s, while failing to segment only 0.1% of cases. Coronary calcium, left atrial volume, and left ventricular mass index are independently associated with all-cause and cardiovascular mortality and significantly improve risk classification compared to identification of abnormalities by a radiologist. This automated approach can be integrated into clinical workflows to improve identification of abnormalities and risk stratification, allowing physicians to improve clinical decision-making. |
| first_indexed | 2024-04-24T16:16:26Z |
| format | Article |
| id | doaj.art-4c588a8fbc88488b95dfaa67b3daf6af |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-04-24T16:16:26Z |
| publishDate | 2024-03-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-4c588a8fbc88488b95dfaa67b3daf6af2024-03-31T11:26:05ZengNature PortfolioNature Communications2041-17232024-03-0115111010.1038/s41467-024-46977-3Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomographyRobert J. H. Miller0Aditya Killekar1Aakash Shanbhag2Bryan Bednarski3Anna M. Michalowska4Terrence D. Ruddy5Andrew J. Einstein6David E. Newby7Mark Lemley8Konrad Pieszko9Serge D. Van Kriekinge10Paul B. Kavanagh11Joanna X. Liang12Cathleen Huang13Damini Dey14Daniel S. Berman15Piotr J. Slomka16Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDivision of Cardiology, University of Ottawa Heart Institute, OttawaDivision of Cardiology, Department of Medicine, Columbia University Irving Medical Center and New York-Presbyterian Hospital, New YorkBritish Heart Foundation Centre for Cardiovascular Science, University of EdinburghDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterDepartments of Medicine (Division of Artificial Intelligence in Medicine), Imaging and Biomedical Sciences Cedars-Sinai Medical CenterAbstract Chest computed tomography is one of the most common diagnostic tests, with 15 million scans performed annually in the United States. Coronary calcium can be visualized on these scans, but other measures of cardiac risk such as atrial and ventricular volumes have classically required administration of contrast. Here we show that a fully automated pipeline, incorporating two artificial intelligence models, automatically quantifies coronary calcium, left atrial volume, left ventricular mass, and other cardiac chamber volumes in 29,687 patients from three cohorts. The model processes chamber volumes and coronary artery calcium with an end-to-end time of ~18 s, while failing to segment only 0.1% of cases. Coronary calcium, left atrial volume, and left ventricular mass index are independently associated with all-cause and cardiovascular mortality and significantly improve risk classification compared to identification of abnormalities by a radiologist. This automated approach can be integrated into clinical workflows to improve identification of abnormalities and risk stratification, allowing physicians to improve clinical decision-making.https://doi.org/10.1038/s41467-024-46977-3 |
| spellingShingle | Robert J. H. Miller Aditya Killekar Aakash Shanbhag Bryan Bednarski Anna M. Michalowska Terrence D. Ruddy Andrew J. Einstein David E. Newby Mark Lemley Konrad Pieszko Serge D. Van Kriekinge Paul B. Kavanagh Joanna X. Liang Cathleen Huang Damini Dey Daniel S. Berman Piotr J. Slomka Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography Nature Communications |
| title | Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography |
| title_full | Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography |
| title_fullStr | Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography |
| title_full_unstemmed | Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography |
| title_short | Predicting mortality from AI cardiac volumes mass and coronary calcium on chest computed tomography |
| title_sort | predicting mortality from ai cardiac volumes mass and coronary calcium on chest computed tomography |
| url | https://doi.org/10.1038/s41467-024-46977-3 |
| work_keys_str_mv | AT robertjhmiller predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT adityakillekar predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT aakashshanbhag predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT bryanbednarski predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT annammichalowska predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT terrencedruddy predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT andrewjeinstein predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT davidenewby predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT marklemley predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT konradpieszko predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT sergedvankriekinge predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT paulbkavanagh predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT joannaxliang predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT cathleenhuang predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT daminidey predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT danielsberman predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography AT piotrjslomka predictingmortalityfromaicardiacvolumesmassandcoronarycalciumonchestcomputedtomography |