World First for Artificial Intelligence To Treat COVID-19 Patients Worldwide

Addenbrooke’s Hospital in Cambridge, along with 20 other hospitals from around the world and healthcare technology leader NVIDIA, have used artificial intelligence (AI) to predict the oxygen needs of Covid patients on a global scale.

The research was fueled by the pandemic and aimed to build an AI tool to predict how much extra oxygen a Covid-19 patient will need in the early days of hospital care, using data from across four continents.

The technique, known as federated learning, used an algorithm to analyze chest X-rays and electronic health data from hospitalized patients with Covid symptoms.

To maintain strict patient confidentiality, patient data was completely anonymized and an algorithm was sent to each hospital so that no data was shared or left their location.

After the algorithm “learned” from the data, the analysis was put together to build an AI tool that could predict the oxygen needs of Covid patients in hospitals around the world.

Published on September 15, 2021 in Nature Medicine, the study called EXAM (for EMR CXR ANl model), is one of the largest and most diverse clinical federated learning studies to date.

To verify the accuracy of EXAM, it was tested in a number of hospitals on five continents, including Addenbrooke’s Hospital. The results showed that it predicted the oxygen needed within 24 hours of a patient’s arrival in the emergency department, with a sensitivity of 95 percent and a specificity of more than 88 percent.

“Federal learning has a transformative power to bring AI innovation to the clinical workflow,” said Professor Fiona Gilbert, who led the research at Cambridge and is an honorary consultant radiologist at Addenbrooke’s Hospital and chair of radiology at the University of Cambridge School of Clinical medicine.

“Our ongoing work with EXAM demonstrates that these types of global collaborations are repeatable and more efficient, enabling us to meet the needs of clinicians to address complex health challenges and future epidemics.”

The study’s lead author, Dr. Ittai Dayan, of Mass General Bingham in the US, where the EXAM algorithm was developed, said:

“Usually in AI development, when you make an algorithm on the data from one hospital, it doesn’t work well in another hospital. By developing the EXAM model using federated learning and objective, multimodal data from different continents, we were able to build a generalizable model that can help primary care physicians worldwide.”

Bringing together collaborators from the Americas, Europe and Asia, the EXAM study took just two weeks of AI learning to make high-quality predictions.

“Federated Learning enabled researchers to collaborate and set a new standard for what we can do globally, leveraging the power of AI,” said Dr. Mona G Flores, Global Head for Medical AI at NVIDIA. “This will advance AI not just for healthcare, but in all industries that want to build robust models without sacrificing privacy.”

The results of approximately 10,000 COVID patients from around the world were analyzed in the study, including 250 who came to Addenbrooke’s Hospital in March/April 2020 in the first wave of the pandemic.

The research was supported by the Cambridge Biomedical Research Center (BRC) of the National Institute for Health Research (NIHR).

Work continues on the EXAM model. Mass General Brigham and the NIHR Cambridge BRC are partnering with NVIDIA Inception startup Rhino Health, co-founded by Dr. Dayan, to conduct prospective studies using EXAM.

Professor Gilbert added: “Creating software to match the performance of our best radiologists is complex, but a truly transformative ambition. The more we can securely integrate data from different sources using federated learning and collaboration, and the more room we have to innovate, the faster academics can achieve those transformative goals.”

Reference: “Federal Learning for Predicting Clinical Outcomes in Patients with COVID-19” by Ittai Dayan, Holger R. Roth, Aoxiao Zhong, Ahmed Harouni, Amilcare Gentili, Anas Z. Abidin, Andrew Liu, Anthony Beardsworth Costa, Bradford J Wood, Chien-Sung Tsai, Chih-Hung Wang, Chun-Nan Hsu, CK Lee, Peiying Ruan, Daguang Xu, Dufan Wu, Eddie Huang, Felipe Campos Kitamura, Griffin Lacey, Gustavo César de Antônio Corradi, Gustavo Nino, Hao -Hsin Shin, Hirofumi Obinata, Hui Ren, Jason C. Crane, Jesse Tetreault, Jiahui Guan, John W. Garrett, Joshua D. Kaggie, Jung Gil Park, Keith Dreyer, Krishna Juluru, Kristopher Kersten, Marcio Aloisio Bezerra Cavalcanti Rockenbach, Marius George Linguraru, Masoom A. Haider, Meena AbdelMaseeh, Nicola Rieke, Pablo F. Damasceno, Pedro Mario Cruz e Silva, Pochuan Wang, Sheng Xu, Shuichi Kawano, Sira Sriswasdi, Soo Young Park, Thomas M. Grist, Varun Buch, Watsamon Jantarabenjakul, Weichung Wang, Won Young Tak, Xiang Li, Xihong Lin, Young Joon Kwon, Abood Quraini, Andrew Feng, Andrew N. Priest, Baris Turkbey, Benjamin Glicksberg, Bernardo Bizzo, Byung Seok Kim, Carlos Tor-Díez, Chia-Cheng Lee, Chia-Jung Hsu, Chin Lin, Chiu-Ling Lai, Christopher P. Hess, Colin Compas, Deepeksha Bhatia, Eric K. Oermann, Evan Leibovitz, Hisashi Sasaki, Hitoshi Mori, Isaac Yang, Jae Ho Sohn, Krishna Nand Keshava Murthy, Li-Chen Fu, Matheus Ribeiro Furtado de Mendonça, Mike Fralick, Min Kyu Kang, Mohammad Adil, Natalie Gangai, Peerapon Vateekul, Pierre Elnajjar, Sarah Hickman, Sharmila Majumdar, Shelley L. McLeod, Sheridan Reed, Stefan Gräf, Stephanie Harmon, Tatsuya Kodama, Thanyawee Puthanakit, Tony Mazzulli, Vitorhai Lima de Lavork Lee, Yuhong Wen, Fiona J. Gilbert, Mona G. Flores, and Quanzheng Li, September 15, 2021, Nature Medicine.
DOI: 10.1038/s41591-021-01506-3