Patient-specific COVID-19 resource utilization prediction using fusion AI model
The strain on healthcare resources brought forth by the recent COVID-19 pandemic has highlighted the need for efficient resource planning and allocation through the prediction of future consumption. Machine learning can predict resource utilization such as the need for hospitalization based on past...
| Main Authors: | , , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Springer Science and Business Media LLC
2021
|
| Online Access: | https://hdl.handle.net/1721.1/130935 |
| _version_ | 1811097265203314688 |
|---|---|
| author | Tariq, Amara Celi, Leo Anthony G. Newsome, Janice M. Purkayastha, Saptarshi Bhatia, Neal Kumar Trivedi, Hari Gichoya, Judy Wawira Banerjee, Imon |
| author2 | Massachusetts Institute of Technology. Institute for Medical Engineering & Science |
| author_facet | Massachusetts Institute of Technology. Institute for Medical Engineering & Science Tariq, Amara Celi, Leo Anthony G. Newsome, Janice M. Purkayastha, Saptarshi Bhatia, Neal Kumar Trivedi, Hari Gichoya, Judy Wawira Banerjee, Imon |
| author_sort | Tariq, Amara |
| collection | MIT |
| description | The strain on healthcare resources brought forth by the recent COVID-19 pandemic has highlighted the need for efficient resource planning and allocation through the prediction of future consumption. Machine learning can predict resource utilization such as the need for hospitalization based on past medical data stored in electronic medical records (EMR). We conducted this study on 3194 patients (46% male with mean age 56.7 (±16.8), 56% African American, 7% Hispanic) flagged as COVID-19 positive cases in 12 centers under Emory Healthcare network from February 2020 to September 2020, to assess whether a COVID-19 positive patient’s need for hospitalization can be predicted at the time of RT-PCR test using the EMR data prior to the test. Five main modalities of EMR, i.e., demographics, medication, past medical procedures, comorbidities, and laboratory results, were used as features for predictive modeling, both individually and fused together using late, middle, and early fusion. Models were evaluated in terms of precision, recall, F1-score (within 95% confidence interval). The early fusion model is the most effective predictor with 84% overall F1-score [CI 82.1–86.1]. The predictive performance of the model drops by 6 % when using recent clinical data while omitting the long-term medical history. Feature importance analysis indicates that history of cardiovascular disease, emergency room visits in the past year prior to testing, and demographic factors are predictive of the disease trajectory. We conclude that fusion modeling using medical history and current treatment data can forecast the need for hospitalization for patients infected with COVID-19 at the time of the RT-PCR test. |
| first_indexed | 2024-09-23T16:56:45Z |
| format | Article |
| id | mit-1721.1/130935 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T16:56:45Z |
| publishDate | 2021 |
| publisher | Springer Science and Business Media LLC |
| record_format | dspace |
| spelling | mit-1721.1/1309352024-06-26T00:06:31Z Patient-specific COVID-19 resource utilization prediction using fusion AI model Tariq, Amara Celi, Leo Anthony G. Newsome, Janice M. Purkayastha, Saptarshi Bhatia, Neal Kumar Trivedi, Hari Gichoya, Judy Wawira Banerjee, Imon Massachusetts Institute of Technology. Institute for Medical Engineering & Science The strain on healthcare resources brought forth by the recent COVID-19 pandemic has highlighted the need for efficient resource planning and allocation through the prediction of future consumption. Machine learning can predict resource utilization such as the need for hospitalization based on past medical data stored in electronic medical records (EMR). We conducted this study on 3194 patients (46% male with mean age 56.7 (±16.8), 56% African American, 7% Hispanic) flagged as COVID-19 positive cases in 12 centers under Emory Healthcare network from February 2020 to September 2020, to assess whether a COVID-19 positive patient’s need for hospitalization can be predicted at the time of RT-PCR test using the EMR data prior to the test. Five main modalities of EMR, i.e., demographics, medication, past medical procedures, comorbidities, and laboratory results, were used as features for predictive modeling, both individually and fused together using late, middle, and early fusion. Models were evaluated in terms of precision, recall, F1-score (within 95% confidence interval). The early fusion model is the most effective predictor with 84% overall F1-score [CI 82.1–86.1]. The predictive performance of the model drops by 6 % when using recent clinical data while omitting the long-term medical history. Feature importance analysis indicates that history of cardiovascular disease, emergency room visits in the past year prior to testing, and demographic factors are predictive of the disease trajectory. We conclude that fusion modeling using medical history and current treatment data can forecast the need for hospitalization for patients infected with COVID-19 at the time of the RT-PCR test. U.S. National Science Foundation, Division Of Electrical, Communication & Cyber Systems (Award 1928481) 2021-06-11T20:31:59Z 2021-06-11T20:31:59Z 2021-06 2020-12 Article http://purl.org/eprint/type/JournalArticle 2398-6352 https://hdl.handle.net/1721.1/130935 Tariq, Amara et al. "Patient-specific COVID-19 resource utilization prediction using fusion AI model." NPJ Digital Medicine 4, 1 (June 2021): 10.1038/s41746-021-00461-0. © 2021 The Author(s) http://dx.doi.org/10.1038/s41746-021-00461-0 NPJ Digital Medicine Creative Commons Attribution 4.0 International license https://creativecommons.org/licenses/by/4.0/ application/pdf Springer Science and Business Media LLC Nature |
| spellingShingle | Tariq, Amara Celi, Leo Anthony G. Newsome, Janice M. Purkayastha, Saptarshi Bhatia, Neal Kumar Trivedi, Hari Gichoya, Judy Wawira Banerjee, Imon Patient-specific COVID-19 resource utilization prediction using fusion AI model |
| title | Patient-specific COVID-19 resource utilization prediction using fusion AI model |
| title_full | Patient-specific COVID-19 resource utilization prediction using fusion AI model |
| title_fullStr | Patient-specific COVID-19 resource utilization prediction using fusion AI model |
| title_full_unstemmed | Patient-specific COVID-19 resource utilization prediction using fusion AI model |
| title_short | Patient-specific COVID-19 resource utilization prediction using fusion AI model |
| title_sort | patient specific covid 19 resource utilization prediction using fusion ai model |
| url | https://hdl.handle.net/1721.1/130935 |
| work_keys_str_mv | AT tariqamara patientspecificcovid19resourceutilizationpredictionusingfusionaimodel AT celileoanthonyg patientspecificcovid19resourceutilizationpredictionusingfusionaimodel AT newsomejanicem patientspecificcovid19resourceutilizationpredictionusingfusionaimodel AT purkayasthasaptarshi patientspecificcovid19resourceutilizationpredictionusingfusionaimodel AT bhatianealkumar patientspecificcovid19resourceutilizationpredictionusingfusionaimodel AT trivedihari patientspecificcovid19resourceutilizationpredictionusingfusionaimodel AT gichoyajudywawira patientspecificcovid19resourceutilizationpredictionusingfusionaimodel AT banerjeeimon patientspecificcovid19resourceutilizationpredictionusingfusionaimodel |