Deep learning-based end-to-end automated stenosis classification and localization on catheter coronary angiography

Abstract

Background Automatic coronary angiography (CAG) assessment may help in faster screening and diagnosis of patients. Current CNN-based vessel-segmentation suffers from sampling imbalance, candidate frame selection, and overfitting; few have shown adequate performance for CAG stenosis classification. We aimed to provide an end-to-end workflow that may solve these problems. Methods A deep learning-based end-to-end workflow was employed as follows: 1) Candidate frame selection from CAG videograms with CNN+LSTM network, 2) Stenosis classification with Inception-v3 using 2 or 3 categories (<25%, >25%, and/or total occlusion) with and without redundancy training, and 3) Stenosis localization with two methods of class activation map (CAM) and anchor-based feature pyramid network (FPN). Overall 13744 frames from 230 studies were used for the stenosis classification training and 4-fold cross-validation for image-, artery-, and per-patient-level. For the stenosis localization training and 4-fold cross-validation, 690 images with >25% stenosis were used. Results Our model achieved an accuracy of 0.85, sensitivity of 0.96, and AUC of 0.86 in per-patient level stenosis classification. Redundancy training was effective to improve classification performance. Stenosis position localization was adequate with better quantitative results in anchor-based FPN model, achieving global-sensitivity for LCA and RCA of 0.68 and 0.70 with mean square error (MSE) values of 39.3 and 37.6 pixels respectively, in the 520 × 520 pixel image. Conclusion A fully-automatic end-to-end deep learning-based workflow that eliminates the vessel extraction and segmentation step was feasible in coronary artery stenosis classification and localization on CAG images.

Automatic construction of coronary artery tree structure based on vessel blood flow tracking

“…In recent years, various deep learning networks have been applied in the auxiliary diagnosis of CAG images. [14][15][16][17][18] However, most of the deep learning models only unilaterally focus on image segmentation, lesion detection, or classification, and rarely consider coronary vessel tree construction. In the field of deep learning, the most popular network structure for medical image segmentation is U-Net.…”

DEEPCADD: A Deep Neural Network for Automatic Detection of Coronary Artery Disease

“…Nesse contexto, o uso de técnicas para o processamento de imagem tem ganhado espac ¸o e provado sua utilidade na análise de DAC [Westra et al 2018]. As ferramentas de triagem recentemente propostas na literature são limitadas pela identificac ¸ão de apenas uma lesão em cada angiografia, falta de validac ¸ão com especialistas [Rodrigues et al 2021, Wua et al 2020] e validac ¸ão com poucos dados [Moon et al 2021, Cong et al 2021, Pang et al 2021. Essas ferramentas tem sido largamente aplicadas no contexto de DAC, porém sem resultados efetivos em cenários reais [Freitas et al 2021a].…”

Coronary Artery Disease Automatic Classification

“…To this end, RCA angiographies are divided into small squares with artery segments to be classified according to the presence or the absence of severe lesions. Emerged methods have achieved accuracy superior to 80% [Cong et al 2021, Moon et al 2021, focused in angiography detection and classification. The main contribution of this study is the segmentation as a preparation module for RCA classification, instead of complete exam evaluation [Moon et al 2021, Cong et al 2021.…”

“…Emerged methods have achieved accuracy superior to 80% [Cong et al 2021, Moon et al 2021, focused in angiography detection and classification. The main contribution of this study is the segmentation as a preparation module for RCA classification, instead of complete exam evaluation [Moon et al 2021, Cong et al 2021. In this work, the pipeline automates the manual frame determination in the coronary angiography assessment to select the best artery segment candidates for severe lesions screening.…”