An AI-powered patient triage platform for future viral outbreaks using COVID-19 as a disease model
Georgia Charkoftaki, Reza Aalizadeh, Álvaro J. Santos‐Neto, Wan Ying Tan, Emily A. Davidson, Varvara Nikolopoulou, Yewei Wang, Brian Thompson, Tristan Furnary, Ying Chen, Elsio A. Wunder, Andreas Coppi, Wade L. Schulz, Akiko Iwasaki, Richard W. Pierce, Charles S. Dela Cruz, Gary V. Désir, Naftali Kaminski, Shelli Farhadian, Kirill Veselkov, Rupak Datta, Melissa Campbell, Νikolaos S. Τhomaidis, Albert I. Ko, Yale IMPACT Study Team, Nathan D. Grubaugh, Allison Nelson, Anne L. Wyllie, Arnau Casanovas‐Massana, Elizabeth B. White, Michael Chiorazzi, Michael Rainone, Rebecca Earnest, Sarah Lapidus, Joseph K. Lim, Maura Nakahata, Ángela Núñez, Denise Shepard, Irene Matos, Yvette Strong, Kelly Anastasio, Kristina Brower, Maxine Kuang, M. Catherine Muenker, Adam J. Moore, Harold Rahming, Laura Glick, Erin Silva, Santos Bermejo, Pavithra Vijayakumar, Bertie Geng, John Fournier, Maksym Minasyan, Sean Bickerton, Melissa Linehan, Patrick Wong, Benjamin Israelow, Anjelica Martin, Tyler Rice, William Khoury-Hanold, Jessica Nouws, David McDonald, Kadi-Ann Rose, Yiyun Cao, Lokesh Sharma, Mikhail Smolgovsky, Abeer Obaid, Giuseppe DeIuliis, Hong‐Jai Park, Nicole Sonnert, Sofia Velazquez, Xiaohua Peng, Michael H. Askenase, Codruta Todeasa, Molly L. Bucklin, Maria Batsu, Alexander J. Robertson, Natasha C. Balkcom, Yicong Liu, Zitong Lin, Coriann E. Dorgay, Ryan Borg, Eréndira C. Di Giuseppe, H. P. Young, Roy S. Herbst, David C. Thompson, Vasilis Vasiliou
Abstract
Abstract Over the last century, outbreaks and pandemics have occurred with disturbing regularity, necessitating advance preparation and large-scale, coordinated response. Here, we developed a machine learning predictive model of disease severity and length of hospitalization for COVID-19, which can be utilized as a platform for future unknown viral outbreaks. We combined untargeted metabolomics on plasma data obtained from COVID-19 patients (n = 111) during hospitalization and healthy controls (n = 342), clinical and comorbidity data (n = 508) to build this patient triage platform, which consists of three parts: (i) the clinical decision tree, which amongst other biomarkers showed that patients with increased eosinophils have worse disease prognosis and can serve as a new potential biomarker with high accuracy (AUC = 0.974), (ii) the estimation of patient hospitalization length with ± 5 days error (R 2 = 0.9765) and (iii) the prediction of the disease severity and the need of patient transfer to the intensive care unit. We report a significant decrease in serotonin levels in patients who needed positive airway pressure oxygen and/or were intubated. Furthermore, 5-hydroxy tryptophan, allantoin, and glucuronic acid metabolites were increased in COVID-19 patients and collectively they can serve as biomarkers to predict disease progression. The ability to quickly identify which patients will develop life-threatening illness would allow the efficient allocation of medical resources and implementation of the most effective medical interventions. We would advocate that the same approach could be utilized in future viral outbreaks to help hospitals triage patients more effectively and improve patient outcomes while optimizing healthcare resources.