A hybrid machine learning/deep learning COVID-19 severity predictive model from CT images and clinical data

TL;DR

This study presents a hybrid machine learning and deep learning model that predicts COVID-19 severity using CT images and clinical data, achieving high accuracy and interpretability for clinical decision support.

Contribution

It introduces a novel combined approach using 3D CNN features, feature selection with Boruta and SHAP, and a CatBoost classifier for accurate severity prediction.

Findings

Achieved a probabilistic AUC of 0.949 on test data.

Integrated CT imaging, laboratory, and clinical data for improved prediction.

Provided case-based SHAP explanations for model interpretability.

Abstract

COVID-19 clinical presentation and prognosis are highly variable, ranging from asymptomatic and paucisymptomatic cases to acute respiratory distress syndrome and multi-organ involvement. We developed a hybrid machine learning/deep learning model to classify patients in two outcome categories, non-ICU and ICU (intensive care admission or death), using 558 patients admitted in a northern Italy hospital in February/May of 2020. A fully 3D patient-level CNN classifier on baseline CT images is used as feature extractor. Features extracted, alongside with laboratory and clinical data, are fed for selection in a Boruta algorithm with SHAP game theoretical values. A classifier is built on the reduced feature space using CatBoost gradient boosting algorithm and reaching a probabilistic AUC of 0.949 on holdout test set. The model aims to provide clinical decision support to medical doctors, with the probability score of belonging to an outcome class and with case-based SHAP interpretation of features importance.