Profiling short-term longitudinal severity progression and associated genes in COVID-19 patients using EHR and single-cell analysis

Abstract Here we propose CovSF, a deep learning model designed to track and forecast short-term severity progression of COVID-19 patients using longitudinal clinical records. The motivation stems from the need for timely medical resource allocation, improved treatment decisions during pandemics, and...

Full description

Saved in:
Bibliographic Details
Main Authors: Seung Hwan Bae, Donghee Kim, Jaeyeon Jang, A-Sol Kim, Soyoon Hwang, Eunkyung Nam, Sohyun Bae, Ji Yeon Lee, Ji Sun Kim, Sang Cheol Kim, Hye-Yeong Jo, Kwangsoo Kim, Inuk Jung, Ki Tae Kwon
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-07793-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849238802541838336
author Seung Hwan Bae
Donghee Kim
Jaeyeon Jang
A-Sol Kim
Soyoon Hwang
Eunkyung Nam
Sohyun Bae
Ji Yeon Lee
Ji Sun Kim
Sang Cheol Kim
Hye-Yeong Jo
Kwangsoo Kim
Inuk Jung
Ki Tae Kwon
author_facet Seung Hwan Bae
Donghee Kim
Jaeyeon Jang
A-Sol Kim
Soyoon Hwang
Eunkyung Nam
Sohyun Bae
Ji Yeon Lee
Ji Sun Kim
Sang Cheol Kim
Hye-Yeong Jo
Kwangsoo Kim
Inuk Jung
Ki Tae Kwon
author_sort Seung Hwan Bae
collection DOAJ
description Abstract Here we propose CovSF, a deep learning model designed to track and forecast short-term severity progression of COVID-19 patients using longitudinal clinical records. The motivation stems from the need for timely medical resource allocation, improved treatment decisions during pandemics, and the understanding of severity progression related immunology. The COVID-19 Severity Forecasting model, CovSF, utilizes 15 clinical features to profile the severity levels of hospital admitted patients and also forecast their severity levels of up to three days ahead. CovSF was trained on a large COVID-19 cohort (n=4,509), achieving an AUROC of 0.92 with 0.85 and 0.89 sensitivity and specificity on an external validation dataset (n=443). The type of oxygen therapy administered was utilized as the target predictive label, which is often used as the severity index. This approach enables the inclusion of a more comprehensive dataset encompassing patients across the full spectrum of severity, rather than restricting the analysis to more narrowly defined outcomes such as ICU admission or mortality. We focused on profiling deteriorating and recovering health conditions, which were validated using patient matched single-cell transcriptomes. Especially, we showed that the immunology significantly differed between the samples during deterioration and recovery, whose severity levels were the same, and thus presenting the importance of longitudinal analysis. We believe that the framework of CovSF can be extended to other respiratory infectious diseases to alleviate the strain of allocating hospital resources, especially in pandemics.
format Article
id doaj-art-ded32ff43c144638bc22a0c7783009e2
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-ded32ff43c144638bc22a0c7783009e22025-08-20T04:01:24ZengNature PortfolioScientific Reports2045-23222025-07-0115111610.1038/s41598-025-07793-xProfiling short-term longitudinal severity progression and associated genes in COVID-19 patients using EHR and single-cell analysisSeung Hwan Bae0Donghee Kim1Jaeyeon Jang2A-Sol Kim3Soyoon Hwang4Eunkyung Nam5Sohyun Bae6Ji Yeon Lee7Ji Sun Kim8Sang Cheol Kim9Hye-Yeong Jo10Kwangsoo Kim11Inuk Jung12Ki Tae Kwon13School of Computer Science and Engineering, Kyungpook National UniversitySeoul National University Hospital Biomedical Research InstituteSchool of Computer Science and Engineering, Kyungpook National UniversityDepartment of Family Medicine, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National UniversityDivision of Infectious Diseases, Department of Internal Medicine, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National UniversityDivision of Infectious Diseases, Department of Internal Medicine, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National UniversityDivision of Infectious Diseases, Department of Internal Medicine, School of Medicine, Kyungpook National University Hospital, Kyungpook National UniversityDivision of Infectious Diseases, Department of Internal Medicine, Keimyung University Dongsan Hospital, Keimyung University School of MedicineDepartment of Medical Information, Kyungpook National University HospitalDivision of Healthcare and Artificial Intelligence, Department of Precision Medicine, Korea National Institute of Health, Korea Disease Control and Prevention AgencyDivision of Healthcare and Artificial Intelligence, Department of Precision Medicine, Korea National Institute of Health, Korea Disease Control and Prevention AgencyDepartment of Transdisciplinary Medicine, Institute of Convergence Medicine with Innovative Technology, Seoul National University HospitalSchool of Computer Science and Engineering, Kyungpook National UniversityDivision of Infectious Diseases, Department of Internal Medicine, Kyungpook National University Chilgok Hospital, School of Medicine, Kyungpook National UniversityAbstract Here we propose CovSF, a deep learning model designed to track and forecast short-term severity progression of COVID-19 patients using longitudinal clinical records. The motivation stems from the need for timely medical resource allocation, improved treatment decisions during pandemics, and the understanding of severity progression related immunology. The COVID-19 Severity Forecasting model, CovSF, utilizes 15 clinical features to profile the severity levels of hospital admitted patients and also forecast their severity levels of up to three days ahead. CovSF was trained on a large COVID-19 cohort (n=4,509), achieving an AUROC of 0.92 with 0.85 and 0.89 sensitivity and specificity on an external validation dataset (n=443). The type of oxygen therapy administered was utilized as the target predictive label, which is often used as the severity index. This approach enables the inclusion of a more comprehensive dataset encompassing patients across the full spectrum of severity, rather than restricting the analysis to more narrowly defined outcomes such as ICU admission or mortality. We focused on profiling deteriorating and recovering health conditions, which were validated using patient matched single-cell transcriptomes. Especially, we showed that the immunology significantly differed between the samples during deterioration and recovery, whose severity levels were the same, and thus presenting the importance of longitudinal analysis. We believe that the framework of CovSF can be extended to other respiratory infectious diseases to alleviate the strain of allocating hospital resources, especially in pandemics.https://doi.org/10.1038/s41598-025-07793-xSeverityProgressionTime courseDeep learningSingle-cell
spellingShingle Seung Hwan Bae
Donghee Kim
Jaeyeon Jang
A-Sol Kim
Soyoon Hwang
Eunkyung Nam
Sohyun Bae
Ji Yeon Lee
Ji Sun Kim
Sang Cheol Kim
Hye-Yeong Jo
Kwangsoo Kim
Inuk Jung
Ki Tae Kwon
Profiling short-term longitudinal severity progression and associated genes in COVID-19 patients using EHR and single-cell analysis
Scientific Reports
Severity
Progression
Time course
Deep learning
Single-cell
title Profiling short-term longitudinal severity progression and associated genes in COVID-19 patients using EHR and single-cell analysis
title_full Profiling short-term longitudinal severity progression and associated genes in COVID-19 patients using EHR and single-cell analysis
title_fullStr Profiling short-term longitudinal severity progression and associated genes in COVID-19 patients using EHR and single-cell analysis
title_full_unstemmed Profiling short-term longitudinal severity progression and associated genes in COVID-19 patients using EHR and single-cell analysis
title_short Profiling short-term longitudinal severity progression and associated genes in COVID-19 patients using EHR and single-cell analysis
title_sort profiling short term longitudinal severity progression and associated genes in covid 19 patients using ehr and single cell analysis
topic Severity
Progression
Time course
Deep learning
Single-cell
url https://doi.org/10.1038/s41598-025-07793-x
work_keys_str_mv AT seunghwanbae profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT dongheekim profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT jaeyeonjang profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT asolkim profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT soyoonhwang profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT eunkyungnam profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT sohyunbae profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT jiyeonlee profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT jisunkim profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT sangcheolkim profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT hyeyeongjo profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT kwangsookim profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT inukjung profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis
AT kitaekwon profilingshorttermlongitudinalseverityprogressionandassociatedgenesincovid19patientsusingehrandsinglecellanalysis