AI Models of Pulmonary Sarcoidosis Detection

Abstract

Sarcoidosis, a multifaceted inflammatory disorder, often involves the ′ lungs, presenting challenges in diagnosis and management. Computed tomography (CT) imaging is pivotal in assessing pulmonary sarcoidosis, yet interpretation can be subjective and variable. This study explores the application of artificial intelligence (AI) models, including Convolutional Neural Networks (CNNs), Residual Networks (ResNets), Recurrent Neural Networks (RNNs), and the Ultralytics model, in automating the detection and classification of pulmonary sarcoidosis lesions on CT scans. Additionally, the research provides a detailed comparative analysis of these AI models, elucidating their strengths and limitations in pulmonary sarcoidosis detection.