Journal d'histologie moléculaire et de physiologie médicale

Utilizing Deep Learning for Comprehensive Lung and Lesion Quantification in Computerized Tomography Amidst Inconsistent Ground Truth

Abstract

Devashish Nath

Computed Tomography (CT) imaging plays a pivotal role in diagnosing, characterizing, predicting outcomes, and tracking disease progression in individuals affected by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Yet, for a consistent and dependable assessment of pulmonary irregularities, precise segmentation and quantification of both the complete lung and lung lesions (anomalies) in chest CT scans of COVID-19 patients are indispensable. Regrettably, the manual segmentation and quantification of extensive datasets can prove time-intensive and yield low levels of agreement both between different observers and within the same observer, even among experienced radiologists.