Explainable AI to identify radiographic features of pulmonary edema

Abstract Summary A deep learning methodology based on an ensemble of segmentation and object detection networks paves the way for an explainable, automated tool to assist in diagnosing and classifying pulmonary edema severity in chest radiographs. Background Pulmonary edema is a leading cause in patients with congestive heart failure (CHF) requiring hospitalization. Assessing the severity of this condition, with radiological imaging, becomes paramount in determining the optimal course of patient care. Purpose This study aimed to develop a deep-learning methodology for the identification of radiographic features associated with pulmonary edema. Materials and Methods This retrospective study utilized a dataset from the MIMIC database comprising 1000 chest X-ray images from 741 patients with suspected pulmonary edema. The images were annotated by an experienced radiologist, who labeled radiographic manifestations of cephalization, Kerley lines, pleural effusion, bat wings, and infiltrate features of edema. The proposed methodology involves two consecutive stages: lung segmentation and edema feature localization. The segmentation stage is implemented using an ensemble of three networks. In the subsequent localization stage, we evaluated eight object detection networks, assessing their performance, with average precision (AP) and mean AP (mAP). Results Effusion, infiltrate, and bat wings features were best detected by the Side-Aware Boundary Localization (SABL) network with corresponding APs of 0.599, 0.395, and 0.926, respectively. Furthermore, SABL achieved the highest overall mAP of 0.568. The Cascade Region Proposal Network (Cascade RPN) network attained the highest AP of 0.417 for Kerley lines and the Probabilistic Anchor Assignment (PAA) network achieved the highest AP of 0.533 for cephalization. Conclusion The proposed methodology, with the application of SABL, Cascade RPN, and PAA detection networks, is accurate and efficient in localizing and identifying pulmonary edema features and is, therefore, a promising diagnostic candidate for interpretable severity assessment of pulmonary edema.