Automated Quantification of Inflamed Lung Regions in Chest CT by UNet++ and SegCaps: A Comparative Analysis in COVID-19 Cases.

During the current COVID-19 pandemic, a high volume of lung imaging has been generated in the aid of the treating clinician. Importantly, lung inflammation severity, associated with the disease outcome, needs to be precisely quantified. Producing consistent and accurate reporting in high-demand scenarios can be a challenge that can compromise patient care with significant inter- or intra-observer variability in quantifying lung inflammation in a chest CT scan. In this backdrop, automated segmentation has recently been attempted using UNet++, a convolutional neural network (CNN), and results comparable to manual methods have been reported. In this paper, we hypothesize that the desired task can be performed with comparable efficiency using capsule networks with fewer parameters that make use of an advanced vector representation of information and dynamic routing. In this paper, we validate this hypothesis using SegCaps, a capsule network, by direct comparison, individual comparison with CT severity score, and comparing the relative effect on a ML(machine learning)-based prognosis model developed elsewhere. We further provide a scenario, where a combination of UNet++ and SegCaps achieves improved performance compared to individual models.

An ML prediction model based on clinical parameters and automated CT scan features for COVID-19 patients.

Outcome prediction for individual patient groups is of paramount importance in terms of selection of appropriate therapeutic options, risk communication to patients and families, and allocating resource through optimum triage. This has become even more necessary in the context of the current COVID-19 pandemic. Widening the spectrum of predictor variables by including radiological parameters alongside the usually utilized demographic, clinical and biochemical ones can facilitate building a comprehensive prediction model. Automation has the potential to build such models with applications to time-critical environments so that a clinician will be able to utilize the model outcomes in real-time decision making at bedside. We show that amalgamation of computed tomogram (CT) data with clinical parameters (CP) in generating a Machine Learning model from 302 COVID-19 patients presenting to an acute care hospital in India could prognosticate the need for invasive mechanical ventilation. Models developed from CP alone, CP and radiologist derived CT severity score and CP with automated lesion-to-lung ratio had AUC of 0.87 (95% CI 0.85-0.88), 0.89 (95% CI 0.87-0.91), and 0.91 (95% CI 0.89-0.93), respectively. We show that an operating point on the ROC can be chosen to aid clinicians in risk characterization according to the resource availability and ethical considerations. This approach can be deployed in more general settings, with appropriate calibrations, to predict outcomes of severe COVID-19 patients effectively.