Automated classification of coronary atherosclerotic plaque in optical frequency domain imaging based on deep learning

Highlights

• A deep learning (DL) model for automated classification of optical frequency domain imaging (OFDI) was developed.

• It can provide accurate categorization of the atherosclerotic coronary plaque type.

• This model can support interventional cardiologists in catheterization laboratory.

Abstract

Background and aims

We developed a deep learning (DL) model for automated atherosclerotic plaque categorization using optical frequency domain imaging (OFDI) and performed quantitative and visual evaluations.

Methods

A total of 1103 histological cross-sections from 45 autopsy hearts were examined to compare the ex vivo OFDI scans. The images were segmented and annotated considering four histological categories: pathological intimal thickening (PIT), fibrous cap atheroma (FA), fibrocalcific plaque (FC), and healed erosion/rupture (HER). The DL model was developed based on pyramid scene parsing network (PSPNet). Given an input image, a convolutional neural network (ResNet50) was used as an encoder to generate feature maps of the last convolutional layer.

Results

For the quantitative evaluation, the mean F-score and IoU values, which are used to evaluate how close the predicted results are to the ground truth, were used. The validation and test dataset had F-score and IoU values of 0.63, 0.49, and 0.66, 0.52, respectively. For the section-level diagnostic accuracy, the areas under the receiver-operating characteristic curve produced by the DL model for FC, PIT, FA, and HER were 0.91, 0.85, 0.86, and 0.86, respectively, and were comparable to those of an expert observer.

Conclusions

DL semantic segmentation of coronary plaques in OFDI images was used as a tool to automatically categorize atherosclerotic plaques using histological findings as the gold standard. The proposed method can support interventional cardiologists in understanding histological properties of plaques.

Introduction

Optical frequency domain imaging (OFDI) and optical coherence tomography (OCT) are widely used as imaging tools for characterizing coronary atherosclerotic plaques in vivo [[1], [2], [3], [4]]. Previous ex vivo studies have reported that OCT can be used with high sensitivity and specificity [5,6] to differentiate between three types of atherosclerotic plaque components (fibrous, fibrocalcific, and lipid-rich) in autopsy specimens. OFDI/OCT can also detect macrophages, necrotic cores, cholesterol crystals, calcified nodules, and intimal vasculature [7,8]. However, the accuracy of OFDI/OCT for histological diagnosis of atherosclerotic lesions depends on the experience of the observer [9]. The accurate interpretation of OFDI scans can be challenging for less experienced observers.

Deep learning (DL) is a field of artificial intelligence that has received significant attention in recent years [10]. Semantic segmentation in DL is the process of partitioning a digital image into multiple segments. This technique, also known as pixel labeling, is used to extract objects by classifying them on a pixel-by-pixel basis. In this study, we hypothesized that pathological classification from OFDI images can be done by the DL-based segmentation model.

We have presented a new computerized technique for atherosclerotic plaque categorization. The main contribution of our work is improvements in the feature extraction stage, wherein the desired features are determined using auto-correlation in a completely automatic process without the need for time-consuming training.