AUTHOR=Holmberg Olle , Lenz Tobias , Koch Valentin , Alyagoob Aseel , Utsch Léa , Rank Andreas , Sabic Emina , Seguchi Masaru , Xhepa Erion , Kufner Sebastian , Cassese Salvatore , Kastrati Adnan , Marr Carsten , Joner Michael , Nicol Philipp TITLE=Histopathology-Based Deep-Learning Predicts Atherosclerotic Lesions in Intravascular Imaging JOURNAL=Frontiers in Cardiovascular Medicine VOLUME=Volume 8 - 2021 YEAR=2021 URL=https://www.frontiersin.org/journals/cardiovascular-medicine/articles/10.3389/fcvm.2021.779807 DOI=10.3389/fcvm.2021.779807 ISSN=2297-055X ABSTRACT=Background: Optical coherence tomography is a powerful modality to assess atherosclerotic lesions, but detecting lesions in high-resolution OCT is challenging and requires expert knowledge. Deep-learning algorithms can be used to automatically identify atherosclerotic lesions, facilitating identification of patients at risk. We trained a deep-learning algorithm (DeepAD) with co-registered, annotated histopathology to detect atherosclerotic lesions in optical coherence tomography (OCT). Methods: Two datasets were used for training DeepAD: (i) a histopathology data set from 7 autopsy cases with 62 OCT frames and co-registered histopathology for high quality manual annotation and (ii) a clinical data set from 51 patients with 222 OCT frames in which manual annotations were based on clinical expertise only. A U-net based deep convolutional neural network (CNN) was employed as an atherosclerotic lesion detection algorithm. Results were analyzed with two metrics: intersection over union (IOU) for segmentation and accuracy of a-line classifications. Results: DeepAD showed good performance regarding the detection of atherosclerotic lesions, with a median IOU of 0.68±0.18 for segmentation and an accuracy of 88% for a-line classification of atherosclerotic tissue. When training the algorithm without histopathology-based annotations, a performance drop of >0.25 IOU was observed. The practical application of DeepAD was evaluated retrospectively in two real-world clinical cases, showing similar performance as compared to expert analysis. Conclusion: Automated detection of atherosclerotic lesions in OCT is improved using a histopathology based deep-learning algorithm, allowing accurate detection in the clinical setting. An automated decision-support tool based on DeepAD could help in risk prediction and guide interventional treatment decisions.