BEHRTDAY: Dynamic Mortality Risk Prediction using Time-Variant COVID-19 Patient Specific Trajectories

Alaleh Azhir; Soheila Talebi; Louis-Henri Merino; Yikuan Li; Thomas Lukasiewicz; Edgar Argulian; Jagat Narula; Borislava Mihaylova

BEHRTDAY: Dynamic Mortality Risk Prediction using Time-Variant COVID-19 Patient Specific Trajectories

AMIA Jt Summits Transl Sci Proc. 2022 May 23:2022:120-129. eCollection 2022.

Authors

Alaleh Azhir¹, Soheila Talebi², Louis-Henri Merino³, Yikuan Li¹, Thomas Lukasiewicz¹, Edgar Argulian², Jagat Narula², Borislava Mihaylova^{1

4}

Affiliations

¹ University of Oxford, Oxford, UK.
² Icahn School of Medicine at Mount Sinai, New York, NY, USA.
³ Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland.
⁴ Queen Mary University of London, London, UK.

PMID: 35854750
PMCID: PMC9285184

Abstract

Incorporating repeated measurements of vitals and laboratory measurements can improve mortality risk-prediction and identify key risk factors in individualized treatment of COVID-19 hospitalized patients. In this observational study, demographic and laboratory data of all admitted patients to 5 hospitals of Mount Sinai Health System, New York, with COVID-19 positive tests between March 1st and June 8th, 2020, were extracted from electronic medical records and compared between survivors and non-survivors. Next day mortality risk of patients was assessed using a transformer-based model BEHRTDAY fitted to patient time series data of vital signs, blood and other laboratory measurements given the entire patients' hospital stay. The study population includes 3699 COVID-19 positive (57% male, median age: 67) patients. This model had a very high average precision score (0.96) and area under receiver operator curve (0.92) for next-day mortality prediction given entire patients' trajectories, and through masking, it learnt each variable's context.