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

Abstract

Purpose: To determine the performance of a chest radiograph (CXR) severity scoring system combined with clinical and laboratory data in predicting the outcome of COVID-19 patients. Materials and Methods: We retrospectively enrolled 301 patients who had reverse transcriptase-polymerase chain reaction (RT-PCR) positive results for COVID-19. CXRs, clinical and laboratory data were collected. A CXR severity scoring system based on a qualitative evaluation by two expert thoracic radiologists was defined. Based on the clinical outcome, the patients were divided into two classes: moderate/mild (patients who did not die or were not intubated) and severe (patients who were intubated and/or died). ROC curve analysis was applied to identify the cutoff point maximizing the Youden index in the prediction of the outcome. Clinical and laboratory data were analyzed through Boruta and Random Forest classifiers. Results: The agreement between the two radiologist scores was substantial (kappa = 0.76).

Telehealth and Artificial Intelligence Insights into Healthcare during the COVID-19 Pandemic

“…Another AI-based model was designed to see how well a chest radiograph performs through scoring of the severity of COVID-19. The model was integrated with laboratory and clinical evidence to predict the outcomes of COVID-19 for infected patients [ 75 ]. The obtained results showed an accuracy of 84% and AUC = 0.82.…”

Chest imaging in patients with acute respiratory failure because of coronavirus disease 2019

“…Many authors proposed the use of severity score indexes for the evaluation of CXR, based on the type and extension of lung alterations, and demonstrated that higher scores correlated with worse clinical conditions and outcomes. More extended consolidations on initial CXR are associated with a higher rate of hospitalization and invasive mechanical ventilation [ 9 , 16 ▪ , 19 , 20 ].…”

“…3a and b). Interestingly, Artificial Intelligence models have been applied to CXRs with the aim of improving diagnostic accuracy and lowering the need of human workforce during the pandemic [9,22].…”

“…Dedicated pathways for COVID-19 patients were created and many departments were turned into sub-intensive or ICU in order to increase the number of available beds [ 8 ]. The relative lack of resources and hospital beds in the very first period of the pandemic created the need for hospitals to establish, which patients to admit with priority, and the selection often depended on an accurate patient evaluation performed in the emergency department [ 9 , 10 ▪ ]. In this setting, chest imaging had a key role in the diagnosis, evaluation, and risk stratification of patients.…”

The role of artificial intelligence in plain chest radiographs interpretation during the Covid-19 pandemic

“…Chest radiographs are insensitive in early and mild Covid-19 infections and are usually used as a first line tool in the triage of patients in the emergency department. 18,20 Nevertheless, chest radiography has an advantage over computed tomography of being readily available, cheaper, portable with a lower radiation dose. They also aid in the assessment of the disease progression and in detecting other diseases contributing to the patients symptoms.…”