Enhanced Classification Performance using Deep Learning Based Segmentation for Pulmonary Embolism Detection in CT Angiography

Abstract

AbstractObjectivesTo develop a deep learning-based algorithm that automatically and accurately classifies patients as either having pulmonary emboli or not in CT pulmonary angiography (CTPA) examinations.Materials and MethodsFor model development, 700 CTPA examinations from 652 patients performed at a single institution between 2014 and 2018 were used, of which 149 examinations contained 1497 PE traced by radiologists. The nnU-Net deep learning-based segmentation framework was trained using 5-fold cross-validation. To enhance classification, we applied logical rules based on PE volume and probability thresholds. External model evaluation was performed in 770 and 34 CTPAs from two independent datasets.ResultsA total of 1483 CTPA examinations were evaluated. In internal cross-validation and test set, the trained model correctly classified 123 of 128 examinations as positive for PE (sensitivity 96.1%; 95% C.I. 91-98%;P< .05) and 521 of 551 as negative (specificity 94.6%; 95% C.I. 92-96%;P< .05). In the first external test dataset, the trained model correctly classified 31 of 32 examinations as positive (sensitivity 96.9%; 95% C.I. 84-99%;P< .05) and 2 of 2 as negative (specificity 100%; 95% C.I. 34-100%;P< .05). In the second external test dataset, the trained model correctly classified 379 of 385 examinations as positive (sensitivity 98.4%; 95% C.I. 97-99%;P< .05) and 346 of 385 as negative (specificity 89.9%; 95% C.I. 86-93%;P< .05).ConclusionOur automatic pipeline achieved beyond state-of-the-art diagnostic performance of PE in CTPA using nnU-Net for segmentation and volume- and probability-based post-processing for classification.Clinical relevance statementThe proposed deep learning algorithm outperforms existing methods for classification of PE in CTPA scans, leveraging the nnU-Net model and including a novel false positive reduction step. This increased diagnostic performance can enable prioritization of patients with PE for rapid review in emergency radiology.Key PointsAn nnU-Net segmentation framework was applied to patient-level classification in CTPA examinations.The proposed algorithm achieved an accuracy of 94.6% with a sensitivity of 96.1% and a specificity of 94.6% in the internal dataset (n=679).The proposed algorithm showed outstanding performance on both internal and two publicly available external testing datasets (AUC, 98.3%; n=1355).