Evaluation of GPT-4 for 10-year cardiovascular risk prediction: Insights from the UK Biobank and KoGES data
Summary: Cardiovascular disease (CVD) remains a pressing global health concern. While traditional risk prediction methods such as the Framingham and American College of Cardiology/American Heart Association (ACC/AHA) risk scores have been widely used in the practice, artificial intelligence (AI), es...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-02-01
|
| Series: | iScience |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2589004224002438 |
| _version_ | 1797319449822363648 |
|---|---|
| author | Changho Han Dong Won Kim Songsoo Kim Seng Chan You Jin Young Park SungA Bae Dukyong Yoon |
| author_facet | Changho Han Dong Won Kim Songsoo Kim Seng Chan You Jin Young Park SungA Bae Dukyong Yoon |
| author_sort | Changho Han |
| collection | DOAJ |
| description | Summary: Cardiovascular disease (CVD) remains a pressing global health concern. While traditional risk prediction methods such as the Framingham and American College of Cardiology/American Heart Association (ACC/AHA) risk scores have been widely used in the practice, artificial intelligence (AI), especially GPT-4, offers new opportunities. Utilizing large scale of multi-center data from 47,468 UK Biobank participants and 5,718 KoGES participants, this study quantitatively evaluated the predictive capabilities of GPT-4 in comparison with traditional models. Our results suggest that the GPT-based score showed commendably comparable performance in CVD prediction when compared to traditional models (AUROC on UKB: 0.725 for GPT-4, 0.733 for ACC/AHA, 0.728 for Framingham; KoGES: 0.664 for GPT-4, 0.674 for ACC/AHA, 0.675 for Framingham). Even with omission of certain variables, GPT-4’s performance was robust, demonstrating its adaptability to data-scarce situations. In conclusion, this study emphasizes the promising role of GPT-4 in predicting CVD risks across varied ethnic datasets, pointing toward its expansive future applications in the medical practice. |
| first_indexed | 2024-03-08T04:07:04Z |
| format | Article |
| id | doaj.art-eb6d0fa27b8a41a6b76eebf94d9e998d |
| institution | Directory Open Access Journal |
| issn | 2589-0042 |
| language | English |
| last_indexed | 2024-03-08T04:07:04Z |
| publishDate | 2024-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | iScience |
| spelling | doaj.art-eb6d0fa27b8a41a6b76eebf94d9e998d2024-02-09T04:48:56ZengElsevieriScience2589-00422024-02-01272109022Evaluation of GPT-4 for 10-year cardiovascular risk prediction: Insights from the UK Biobank and KoGES dataChangho Han0Dong Won Kim1Songsoo Kim2Seng Chan You3Jin Young Park4SungA Bae5Dukyong Yoon6Department of Biomedical Systems Informatics, Yonsei University College of Medicine, Yongin, Republic of KoreaDepartment of Biomedical Systems Informatics, Yonsei University College of Medicine, Yongin, Republic of KoreaDepartment of Biomedical Systems Informatics, Yonsei University College of Medicine, Yongin, Republic of KoreaDepartment of Biomedical Systems Informatics, Yonsei University College of Medicine, Yongin, Republic of Korea; Institute for Innovation in Digital Healthcare, Severance Hospital, Seoul, Republic of KoreaCenter for Digital Health, Yongin Severance Hospital, Yonsei University Health System, Yongin, Republic of Korea; Department of Psychiatry, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, Republic of Korea; Institute of Behavioral Science in Medicine, Yonsei University College of Medicine, Yonsei University Health System, Seoul, Republic of KoreaCenter for Digital Health, Yongin Severance Hospital, Yonsei University Health System, Yongin, Republic of Korea; Department of Cardiology, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, Republic of Korea; Corresponding authorDepartment of Biomedical Systems Informatics, Yonsei University College of Medicine, Yongin, Republic of Korea; Institute for Innovation in Digital Healthcare, Severance Hospital, Seoul, Republic of Korea; Center for Digital Health, Yongin Severance Hospital, Yonsei University Health System, Yongin, Republic of Korea; Corresponding authorSummary: Cardiovascular disease (CVD) remains a pressing global health concern. While traditional risk prediction methods such as the Framingham and American College of Cardiology/American Heart Association (ACC/AHA) risk scores have been widely used in the practice, artificial intelligence (AI), especially GPT-4, offers new opportunities. Utilizing large scale of multi-center data from 47,468 UK Biobank participants and 5,718 KoGES participants, this study quantitatively evaluated the predictive capabilities of GPT-4 in comparison with traditional models. Our results suggest that the GPT-based score showed commendably comparable performance in CVD prediction when compared to traditional models (AUROC on UKB: 0.725 for GPT-4, 0.733 for ACC/AHA, 0.728 for Framingham; KoGES: 0.664 for GPT-4, 0.674 for ACC/AHA, 0.675 for Framingham). Even with omission of certain variables, GPT-4’s performance was robust, demonstrating its adaptability to data-scarce situations. In conclusion, this study emphasizes the promising role of GPT-4 in predicting CVD risks across varied ethnic datasets, pointing toward its expansive future applications in the medical practice.http://www.sciencedirect.com/science/article/pii/S2589004224002438Health sciencesMedicineHealth informaticsCardiovascular medicineHealth technologyArtificial intelligence |
| spellingShingle | Changho Han Dong Won Kim Songsoo Kim Seng Chan You Jin Young Park SungA Bae Dukyong Yoon Evaluation of GPT-4 for 10-year cardiovascular risk prediction: Insights from the UK Biobank and KoGES data iScience Health sciences Medicine Health informatics Cardiovascular medicine Health technology Artificial intelligence |
| title | Evaluation of GPT-4 for 10-year cardiovascular risk prediction: Insights from the UK Biobank and KoGES data |
| title_full | Evaluation of GPT-4 for 10-year cardiovascular risk prediction: Insights from the UK Biobank and KoGES data |
| title_fullStr | Evaluation of GPT-4 for 10-year cardiovascular risk prediction: Insights from the UK Biobank and KoGES data |
| title_full_unstemmed | Evaluation of GPT-4 for 10-year cardiovascular risk prediction: Insights from the UK Biobank and KoGES data |
| title_short | Evaluation of GPT-4 for 10-year cardiovascular risk prediction: Insights from the UK Biobank and KoGES data |
| title_sort | evaluation of gpt 4 for 10 year cardiovascular risk prediction insights from the uk biobank and koges data |
| topic | Health sciences Medicine Health informatics Cardiovascular medicine Health technology Artificial intelligence |
| url | http://www.sciencedirect.com/science/article/pii/S2589004224002438 |
| work_keys_str_mv | AT changhohan evaluationofgpt4for10yearcardiovascularriskpredictioninsightsfromtheukbiobankandkogesdata AT dongwonkim evaluationofgpt4for10yearcardiovascularriskpredictioninsightsfromtheukbiobankandkogesdata AT songsookim evaluationofgpt4for10yearcardiovascularriskpredictioninsightsfromtheukbiobankandkogesdata AT sengchanyou evaluationofgpt4for10yearcardiovascularriskpredictioninsightsfromtheukbiobankandkogesdata AT jinyoungpark evaluationofgpt4for10yearcardiovascularriskpredictioninsightsfromtheukbiobankandkogesdata AT sungabae evaluationofgpt4for10yearcardiovascularriskpredictioninsightsfromtheukbiobankandkogesdata AT dukyongyoon evaluationofgpt4for10yearcardiovascularriskpredictioninsightsfromtheukbiobankandkogesdata |