Explainable Machine Learning for Early Assessment of COVID-19 Risk Prediction in Emergency Departments

Abstract

Between January and October of 2020, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus has infected more than 34 million persons in a worldwide pandemic leading to over one million deaths worldwide (data from the Johns Hopkins University). Since the virus begun to spread, emergency departments were busy with COVID-19 patients for whom a quick decision regarding in-or outpatient care was required. The virus can cause characteristic abnormalities in chest radiographs (CXR), but, due to the low sensitivity of CXR, additional variables and criteria are needed to accurately predict risk. Here, we describe a computerized system primarily aimed at extracting the most relevant radiological, clinical, and laboratory variables for improving patient risk prediction, and secondarily at presenting an explainable machine learning system, which may provide simple decision criteria to be used by clinicians as a support for assessing patient risk. To achieve robust and reliable variable selection, Boruta and Random Forest (RF) are combined in a 10-fold cross-validation scheme to produce a variable importance estimate not biased by the presence of surrogates. The most important variables are then selected to train a RF classifier, whose rules may be extracted, simplified, and pruned to finally build an associative tree, particularly appealing for its simplicity. Results show that the radiological score automatically computed through a neural network is highly correlated with the score computed by radiologists, and that laboratory variables, together with the number of comorbidities, aid risk prediction. The prediction performance of our approach was compared to that that of generalized linear models and shown to be effective and robust. The proposed machine learning-based computational system can be easily deployed and used in emergency departments for rapid and accurate risk prediction in COVID-19 patients.

Volume-of-Interest Aware Deep Neural Networks for Rapid Chest CT-Based COVID-19 Patient Risk Assessment

“…Unfortunately, though a great deal of research work has been devoted to the development of methods for automated COVID-19 diagnosis [ 18 ], severity scoring or prognosis prediction, and outcome prediction [ 8 , 13 ], we are still far from reaching a solution. That is due to the lack of a unified and anonymized, multi-device and multi-ethnicity, appropriately annotated and shareable datasets containing enough samples to ensure a robust model validation against the COVID-19 disease variability.…”

“…In times of crisis, such as the current COVID-19 pandemic, rapid and efficient patient diagnosis and prognosis assessment would highly improve patient care and reduce mortality rate by eliminating the time intervals between Emergency Department (ED) arrival and hospitalization. For a quick and precise patient risk assessment, Computed Tomography (CT) has been described as an important diagnostic tool [ 5 , 6 ], given its capability of reducing RT-PCR false negative results [ 7 ] and its superior sensitivity compared to chest X-ray [ 8 ]. Indeed, CT quantification of pneumonia lesions can timely and non-invasively predict the progression to severe illness [ 9 ], providing a promising prognostic indicator for clinical management of COVID-19.…”

LungINFseg: Segmenting COVID-19 Infected Regions in Lung CT Images Based on a Receptive-Field-Aware Deep Learning Framework

“…Additional features of COVID-19, such as pleural/septal thickening, subpleural involvement, and bronchiectasis, can be noticed in the later stages of the disease. It is worth noting that some related works, such as [ 5 , 6 ], suggest the use of chest radiographs (CXR) due to their widespread availability and portability, non-invasiveness, and faster acquisition and visual analysis. However, CTs have higher accuracy than CXR and allow diminishing the false negative errors from repeated swab analysis [ 7 ].…”

Artificial Intelligence in Predicting Clinical Outcome in COVID-19 Patients from Clinical, Biochemical and a Qualitative Chest X-Ray Scoring System

“…The imputations obtained were compared to the 100 imputations computed by missForest, [21][22][23] which applies Random Forests (RF) in a similar way to RF-mice. To choose the best imputation model, as detailed in Casiraghi et al, 24 this study used the Wilcoxon rank-sum test (p-value < 0.05) to statistically compare the between-imputation-variances obtained with increasing imputation runs (from 1 to 100). This showed that missForest was the most stable imputer.…”

“…Samples below the CXR cut-off point were considered as moderate/mild patients, while those above the CXR cut-point were further analyzed to select the most important variables for a more precise outcome prediction. To this aim, Boruta algorithm 23,[26][27][28][29] used an internal 5-fold crossvalidation as detailed by Casiraghi et al 24 3. Selected variables were used to train an RF, which was then pruned and simplified to create a simple associative tree 30,31 and to finally estimate the importance of the variables.…”