An Interpretable Machine Learning Framework for Accurate Severe vs Non-severe COVID-19 Clinical Type Classification

Abstract

Effectively and efficiently diagnosing COVID-19 patients with accurate clinical type is essential to achieve optimal outcomes of the patients as well as reducing the risk of overloading the healthcare system. Currently, severe and non-severe COVID-19 types are differentiated by only a few clinical features, which do not comprehensively characterize complicated pathological, physiological, and immunological responses to SARS-CoV-2 invasion in different types. In this study, we recruited 214 confirmed COVID-19 patients in non-severe and 148 in severe type, from Wuhan, China. The patients' comorbidity and symptoms (26 features), and blood biochemistry (26 features) upon admission were acquired as two input modalities. Exploratory analyses demonstrated that these features differed substantially between two clinical types. Machine learning random forest (RF) models using features in each modality were developed and validated to classify COVID-19 clinical types. Using comorbidity/symptom and biochemistry as input independently, RF models achieved >90% and >95% predictive accuracy, respectively. Input features' importance based on Gini impurity were further evaluated and top five features from each modality were identified (age, hypertension, cardiovascular disease, gender, diabetes; D-Dimer, hsTNI, neutrophil, IL-6, and LDH). Combining top 10 multimodal features, RF model achieved >99% predictive accuracy. These findings shed light on how the human body reacts to SARS-CoV-2 invasion as a unity and provide insights on effectively evaluating COVID-19 patient's severity and developing treatment plans accordingly. We suggest that symptoms and comorbidities can be used as an initial screening tool for triaging, while biochemistry and features combined are applied when accuracy is the priority.

Predicting the disease outcome in COVID-19 positive patients through Machine Learning: a retrospective cohort study with Brazilian data

“…Prediction models of the prognosis for a given disease have the main objective of supporting the physician's decision-making about what is the best measure of patient referral, assisting in the screening of patients at high risk of progressing to severe disease. Artificial intelligence models aiming to identify risk factors for prognostic prediction of severe COVID-19 have been developed using age, clinical characteristics, laboratory tests and chest imaging [30,31,32,29,33,34].…”

“…The addition of biochemical data to symptoms/comorbidities achieved > 99% predictive accuracy. Therefore, it was suggested that symptoms and comorbidities can be used in an initial screening and the biochemical data inclusion could predict the severity degree and assist in the development of treatment plans [29].…”

Accurately Differentiating Between Patients With COVID-19, Patients With Other Viral Infections, and Healthy Individuals: Multimodal Late Fusion Learning Approach

“…An advantage that the RF model had over SVM and kNN models was that it had relatively clearer interpretability, especially when interpreting feature importance. After developing the RF model based on the training set, we were able to rank the importance of input features based on their corresponding Gini impurity score from the RF model [ 40 , 41 ]. It should be noted that only the training set was used to compute Gini impurity, not the test set.…”

Applications of artificial intelligence in battling against covid-19: A literature review

“…The authors use machine learning algorithms to provide assay design for detection of 67 viral species and subspecies of SARS-CoV-2. In [61] , random forest models are used to classify the covid-19 patients.…”