Federated Learning used for predicting outcomes in SARS-COV-2 patients

‘Federated Learning’ (FL) is a method to train Artificial Intelligence (AI) models with data from multiple sources while maintaining anonymity of the data thus removing many barriers to data sharing. During the SARS-COV-2 pandemic, 20 institutes collaborated on a healthcare FL study to predict futur...

Descripción completa

Detalles Bibliográficos
Autores principales: Flores, Mona, Dayan, Ittai, Roth, Holger, Zhong, Aoxiao, Harouni, Ahmed, Gentili, Amilcare, Abidin, Anas, Liu, Andrew, Costa, Anthony, Wood, Bradford, Tsai, Chien-Sung, Wang, Chih-Hung, Hsu, Chun-Nan, Lee, CK, Ruan, Colleen, Xu, Daguang, Wu, Dufan, Huang, Eddie, Kitamura, Felipe, Lacey, Griffin, César de Antônio Corradi, Gustavo, Shin, Hao-Hsin, Obinata, Hirofumi, Ren, Hui, Crane, Jason, Tetreault, Jesse, Guan, Jiahui, Garrett, John, Park, Jung Gil, Dreyer, Keith, Juluru, Krishna, Kersten, Kristopher, Bezerra Cavalcanti Rockenbach, Marcio Aloisio, Linguraru, Marius, Haider, Masoom, AbdelMaseeh, Meena, Rieke, Nicola, Damasceno, Pablo, Cruz e Silva, Pedro Mario, Wang, Pochuan, Xu, Sheng, Kawano, Shuichi, Sriswasdi, Sira, Park, Soo Young, Grist, Thomas, Buch, Varun, Jantarabenjakul, Watsamon, Wang, Weichung, Tak, Won Young, Li, Xiang, Lin, Xihong, Kwon, Fred, Gilbert, Fiona, Kaggie, Josh, Li, Quanzheng, Quraini, Abood, Feng, Andrew, Priest, Andrew, Turkbey, Baris, Glicksberg, Benjamin, Bizzo, Bernardo, Kim, Byung Seok, Tor-Diez, Carlos, Lee, Chia-Cheng, Hsu, Chia-Jung, Lin, Chin, Lai, Chiu-Ling, Hess, Christopher, Compas, Colin, Bhatia, Deepi, Oermann, Eric, Leibovitz, Evan, Sasaki, Hisashi, Mori, Hitoshi, Yang, Isaac, Sohn, Jae Ho, Keshava Murthy, Krishna Nand, Fu, Li-Chen, Furtado de Mendonça, Matheus Ribeiro, Fralick, Mike, Kang, Min Kyu, Adil, Mohammad, Gangai, Natalie, Vateekul, Peerapon, Elnajjar, Pierre, Hickman, Sarah, Majumdar, Sharmila, McLeod, Shelley, Reed, Sheridan, Graf, Stefan, Harmon, Stephanie, Kodama, Tatsuya, Puthanakit, Thanyawee, Mazzulli, Tony, de Lima Lavor, Vitor, Rakvongthai, Yothin, Lee, Yu Rim, Wen, Yuhong
Formato: Online Artículo Texto
Lenguaje:English
Publicado: American Journal Experts 2021
Materias:
Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7805458/
https://www.ncbi.nlm.nih.gov/pubmed/33442676
http://dx.doi.org/10.21203/rs.3.rs-126892/v1
Descripción
Sumario:‘Federated Learning’ (FL) is a method to train Artificial Intelligence (AI) models with data from multiple sources while maintaining anonymity of the data thus removing many barriers to data sharing. During the SARS-COV-2 pandemic, 20 institutes collaborated on a healthcare FL study to predict future oxygen requirements of infected patients using inputs of vital signs, laboratory data, and chest x-rays, constituting the “EXAM” (EMR CXR AI Model) model. EXAM achieved an average Area Under the Curve (AUC) of over 0.92, an average improvement of 16%, and a 38% increase in generalisability over local models. The FL paradigm was successfully applied to facilitate a rapid data science collaboration without data exchange, resulting in a model that generalised across heterogeneous, unharmonized datasets. This provided the broader healthcare community with a validated model to respond to COVID-19 challenges, as well as set the stage for broader use of FL in healthcare.

Ejemplares similares