Comparative assessment of machine learning and physician-certified verbal autopsy in COVID-19 mortality

Abstract

The aim of this study is to ascertain the underlying causes of death in deceased individuals by employing various classification methods. This investigation uses the verbal autopsy approach, which relies on narrative information, to determine the causes of death in the deceased population. To achieve this objective, the study uses a range of machine learning models, including Random Forest, Decision Tree, K-Nearest Neighbors (KNN), Support Vector Classifier (SVC), and XGBoost. Data for the study was collected from 11 rural areas within the Varanasi district, using a well-structured questionnaire. Verbal Autopsy form captured demographic details such as the age and gender of the deceased individuals, as well as information about the symptoms experienced by the deceased individuals and any coexisting medical conditions they may have had. The study aims to draw conclusions about the accuracy of cause-of-death assignments for deceased individuals using various classifier approaches, providing valuable insights for healthcare professionals.

Ähnliche Arbeiten