TY - JOUR
T1 - Crowd-sourced machine learning prediction of long COVID using data from the National COVID Cohort Collaborative
AU - Long COVID Computational Challenge Participants
AU - N3C Consortium
AU - Bergquist, Timothy
AU - Loomba, Johanna
AU - Pfaff, Emily
AU - Xia, Fangfang
AU - Zhao, Zixuan
AU - Zhu, Yitan
AU - Mitchell, Elliot
AU - Bhattacharya, Biplab
AU - Shetty, Gaurav
AU - Munia, Tamanna
AU - Delong, Grant
AU - Tariq, Adbul
AU - Butzin-Dozier, Zachary
AU - Ji, Yunwen
AU - Li, Haodong
AU - Coyle, Jeremy
AU - Shi, Seraphina
AU - Philips, Rachael V.
AU - Mertens, Andrew
AU - Pirracchio, Romain
AU - van der Laan, Mark
AU - Colford, John M.
AU - Hubbard, Alan
AU - Gao, Jifan
AU - Chen, Guanhua
AU - Velingker, Neelay
AU - Li, Ziyang
AU - Wu, Yinjun
AU - Stein, Adam
AU - Huang, Jiani
AU - Dai, Zongyu
AU - Long, Qi
AU - Naik, Mayur
AU - Holmes, John
AU - Mowery, Danielle
AU - Wong, Eric
AU - Parekh, Ravi
AU - Getzen, Emily
AU - Hightower, Jake
AU - Blase, Jennifer
AU - Aggarwal, Ataes
AU - Agor, Joseph
AU - Al-Amery, Amera
AU - Aminu, Oluwatobiloba
AU - Anand, Adit
AU - Antonescu, Corneliu
AU - Arora, Mehak
AU - Asaduzzaman, Sayed
AU - Asmussen, Tanner
AU - Baghbanzadeh, Mahdi
AU - Baker, Frazier
AU - Bangert, Bridget
AU - Bekhet, Laila
AU - Blase, Jenny
AU - Caffo, Brian
AU - Chang, Hao
AU - Chen, Zeyuan
AU - Chen, Jiandong
AU - Chiang, Jeffrey
AU - Cho, Peter
AU - Cockrell, Robert
AU - Combs, Parker
AU - Crosby, Ciara
AU - Dai, Ran
AU - Danesharasteh, Anseh
AU - Yildirim, Elif
AU - Demilt, Ryan
AU - Deng, Kaiwen
AU - Dey, Sanjoy
AU - Dhamdhere, Rohan
AU - Dickson, Andrew
AU - Dijour, Phoebe
AU - Dinh, Dong
AU - Dixon, Richard
AU - Domi, Albi
AU - Dutta, Souradeep
AU - Elizondo, Mirna
AU - Ertem, Zeynep
AU - Feuerwerker, Solomon
AU - Fliss, Danica
AU - Fowler, Jennifer
AU - Fu, Sunyang
AU - Gardner, Kelly
AU - Getty, Neil
AU - Ghalwash, Mohamed
AU - Gloster, Logan
AU - Greer, Phil
AU - Guan, Yuanfang
AU - Ham, Colby
AU - Hanoudi, Samer
AU - Harper, Jeremy
AU - Hendrix, Nathaniel
AU - Hershkovich, Leeor
AU - Hu, Junjie
AU - Huang, Yu
AU - Huang, Tongtong
AU - Hur, Junguk
AU - Isgut, Monica
AU - Ismail, Hamid
AU - Izmirlian, Grant
AU - Jang, Kuk
AU - Jemiyo, Christianah
AU - Jeong, Hayoung
AU - Ji, Xiayan
AU - Jiang, Ming
AU - Jiang, Sihang
AU - Jiang, Xiaoqian
AU - Jiang, Yuye
AU - Johnson, Akin
AU - Analyst, Zach
AU - Kapse, Saarthak
AU - Kartoun, Uri
AU - KC, Dukka
AU - Fard, Zahra
AU - Kosfeld, Tim
AU - Krichevsky, Spencer
AU - Kuo, Mike
AU - Larie, Dale
AU - Lederer, Lauren
AU - Leng, Shan
AU - Li, Hongyang
AU - Li, Jianfu
AU - Li, Tiantian
AU - Liang, Xinwen
AU - Liang, Hengyue
AU - Liu, Feifan
AU - Liu, Daniel
AU - Luo, Gang
AU - Madduri, Ravi
AU - Madhira, Vithal
AU - Mani, Shivali
AU - Mansourifard, Farzaneh
AU - Matson, Robert
AU - Metsis, Vangelis
AU - Meyer, Pablo
AU - Mikhailova, Catherine
AU - Miller, Dante
AU - Milo, Christopher
AU - Modanwal, Gourav
AU - Moore, Ronald
AU - Morgenthaler, David
AU - Musal, Rasim
AU - Nalawade, Vinit
AU - Narain, Rohan
AU - Narendrula, Saideep
AU - Obiri, Alena
AU - Okawa, Satoshi
AU - Okechukwu, Chima
AU - Olorunnisola, Toluwanimi
AU - Ossowski, Tim
AU - Parekh, Harsh
AU - Park, Jean
AU - Patel, Saaya
AU - Patterson, Jason
AU - Paul, Chetan
AU - Peng, Le
AU - Perkins, Diana
AU - Pokharel, Suresh
AU - Poplavskiy, Dmytro
AU - Pryor, Zach
AU - Pungitore, Sarah
AU - Qin, Hong
AU - Rababa, Salahaldeen
AU - Rahman, Mahbubur
AU - Rahmani, Elior
AU - Rahnavard, Gholamali
AU - Raihan, Md
AU - Rajendran, Suraj
AU - Ravichandran, Sarangan
AU - Reddy, Chandan
AU - Reyes, Abel
AU - Roghanizad, Ali
AU - Rouffa, Sean
AU - Ruan, Xiaoyang
AU - Saha, Arpita
AU - Sawant, Sahil
AU - Schiaffino, Melody
AU - Seira, Diego
AU - Sengupta, Saurav
AU - Shalaev, Ruslan
AU - Shinguyen, Linh
AU - Singh, Karnika
AU - Sinha, Soumya
AU - Socia, Damien
AU - Stalians, Halen
AU - Stavropoulos, Charalambos
AU - Strube, Jan
AU - Subramanian, Devika
AU - Sun, Jiehuan
AU - Sun, Ju
AU - Sun, Chengkun
AU - Sundararajan, Prathic
AU - Talebi, Salmonn
AU - Tawiah, Edward
AU - Tesic, Jelena
AU - Thiess, Mikaela
AU - Tian, Raymond
AU - Torre-Healy; Ming-Tse Tsai, Luke
AU - Payne, Philip
AU - Wilcox, Adam
N1 - Publisher Copyright:
© 2024 The Authors
PY - 2024/10
Y1 - 2024/10
N2 - Background: While many patients seem to recover from SARS-CoV-2 infections, many patients report experiencing SARS-CoV-2 symptoms for weeks or months after their acute COVID-19 ends, even developing new symptoms weeks after infection. These long-term effects are called post-acute sequelae of SARS-CoV-2 (PASC) or, more commonly, Long COVID. The overall prevalence of Long COVID is currently unknown, and tools are needed to help identify patients at risk for developing long COVID. Methods: A working group of the Rapid Acceleration of Diagnostics-radical (RADx-rad) program, comprised of individuals from various NIH institutes and centers, in collaboration with REsearching COVID to Enhance Recovery (RECOVER) developed and organized the Long COVID Computational Challenge (L3C), a community challenge aimed at incentivizing the broader scientific community to develop interpretable and accurate methods for identifying patients at risk of developing Long COVID. From August 2022 to December 2022, participants developed Long COVID risk prediction algorithms using the National COVID Cohort Collaborative (N3C) data enclave, a harmonized data repository from over 75 healthcare institutions from across the United States (U.S.). Findings: Over the course of the challenge, 74 teams designed and built 35 Long COVID prediction models using the N3C data enclave. The top 10 teams all scored above a 0.80 Area Under the Receiver Operator Curve (AUROC) with the highest scoring model achieving a mean AUROC of 0.895. Included in the top submission was a visualization dashboard that built timelines for each patient, updating the risk of a patient developing Long COVID in response to clinical events. Interpretation: As a result of L3C, federal reviewers identified multiple machine learning models that can be used to identify patients at risk for developing Long COVID. Many of the teams used approaches in their submissions which can be applied to future clinical prediction questions. Funding: Research reported in this RADx® Rad publication was supported by the National Institutes of Health. Timothy Bergquist, Johanna Loomba, and Emily Pfaff were supported by Axle Subcontract: NCATS-STSS-P00438.
AB - Background: While many patients seem to recover from SARS-CoV-2 infections, many patients report experiencing SARS-CoV-2 symptoms for weeks or months after their acute COVID-19 ends, even developing new symptoms weeks after infection. These long-term effects are called post-acute sequelae of SARS-CoV-2 (PASC) or, more commonly, Long COVID. The overall prevalence of Long COVID is currently unknown, and tools are needed to help identify patients at risk for developing long COVID. Methods: A working group of the Rapid Acceleration of Diagnostics-radical (RADx-rad) program, comprised of individuals from various NIH institutes and centers, in collaboration with REsearching COVID to Enhance Recovery (RECOVER) developed and organized the Long COVID Computational Challenge (L3C), a community challenge aimed at incentivizing the broader scientific community to develop interpretable and accurate methods for identifying patients at risk of developing Long COVID. From August 2022 to December 2022, participants developed Long COVID risk prediction algorithms using the National COVID Cohort Collaborative (N3C) data enclave, a harmonized data repository from over 75 healthcare institutions from across the United States (U.S.). Findings: Over the course of the challenge, 74 teams designed and built 35 Long COVID prediction models using the N3C data enclave. The top 10 teams all scored above a 0.80 Area Under the Receiver Operator Curve (AUROC) with the highest scoring model achieving a mean AUROC of 0.895. Included in the top submission was a visualization dashboard that built timelines for each patient, updating the risk of a patient developing Long COVID in response to clinical events. Interpretation: As a result of L3C, federal reviewers identified multiple machine learning models that can be used to identify patients at risk for developing Long COVID. Many of the teams used approaches in their submissions which can be applied to future clinical prediction questions. Funding: Research reported in this RADx® Rad publication was supported by the National Institutes of Health. Timothy Bergquist, Johanna Loomba, and Emily Pfaff were supported by Axle Subcontract: NCATS-STSS-P00438.
KW - COVID-19
KW - Community challenge
KW - Evaluation
KW - Long COVID
KW - Machine learning
KW - PASC
UR - http://www.scopus.com/inward/record.url?scp=85204559803&partnerID=8YFLogxK
U2 - 10.1016/j.ebiom.2024.105333
DO - 10.1016/j.ebiom.2024.105333
M3 - Article
C2 - 39321500
AN - SCOPUS:85204559803
SN - 2352-3964
VL - 108
JO - EBioMedicine
JF - EBioMedicine
M1 - 105333
ER -