Explainable Machine-Learning Models for COVID-19 Prognosis Prediction Using Clinical, Laboratory and Radiomic Features

Abstract

The SARS-CoV-2 virus pandemic had devastating effects on various aspects of life: clinical cases, ranging from mild to severe, can lead to lung failure and to death. Due to the high incidence, datadriven models can support physicians in patient management. The explainability and interpretability of machine-learning models are mandatory in clinical scenarios. In this work, clinical, laboratory and radiomic features were used to train machine-learning models for COVID-19 prognosis prediction. Using Explainable AI algorithms, a multi-level explainable method was proposed taking into account the developer and the involved stakeholder (physician, and patient) perspectives. A total of 1023 radiomic features were extracted from 1589 Chest X-Ray images (CXR), combined with 38 clinical/laboratory features. After the preprocessing and selection phases, 40 CXR radiomic features and 23 clinical/laboratory features were used to train Support Vector Machine and Random Forest classifiers exploring three feature selection strategies. The combination of both radiomic, and clinical/laboratory features enabled higher performance in the resulting models. The intelligibility of the used features allowed us to validate the models' clinical findings. According to the medical literature, LDH, PaO2 and CRP were the most predictive laboratory features. Instead, ZoneEntropy and HighGrayLevelZoneEmphasis -indicative of the heterogeneity/uniformity of lung texture -were the most discriminating radiomic features. Our best predictive model, exploiting the Random Forest classifier and a signature composed of clinical, laboratory and radiomic features, achieved AUC=0.819, accuracy=0.733, specificity=0.705, and sensitivity=0.761 in the test set. The model, including a multi-level explainability, allows us to make strong clinical assumptions, confirmed by the literature insights.INDEX TERMS Chest X-ray images, clinical and laboratory features, COVID-19 prognosis, explainable AI, machine learning classifiers, predictive models, radiomic features.

Shallow and deep learning classifiers in medical image analysis

“…Subsequently, a cross-validation strategy is often employed exclusively in the training set. This cross-validation approach, consisting of dividing the training set into subsets (called folds), and in each round select one of these folds as the validation set and the others as the training set, is used for both training and fine-tuning the model, and, ultimately, the model’s performance is assessed on the dedicated test set [ 18 ]. In the case of very small datasets (about less than 100 samples), the leave-one-out method is typically employed [ 19 , 20 ].…”

“…Conversely, a local explanation focuses on elucidating the system’s decision for a particular instance, such as a patient. This approach allows for a detailed examination of the model’s findings and facilitates clinical validation and comparisons with existing medical literature [ 18 ]. These considerations carry significant ethical, legal, and trust-related implications.…”

Leveraging Radiomics and Genetic Algorithms to Improve Lung Infection Diagnosis in X-Ray Images Using Machine Learning

“…F. Prinzi et.al proposed the significance of employing explainable AI algorithms in clinical settings to ensure the interpretability of predictive models for COVID-19 prognosis. This emphasis on interpretability facilitates informed clinical decision-making [29]. De et al [30] introduced a widely practiced intervention for modifying cardiac health, highlighting the varied effects of physical activity on older adults.…”