Automatic Plaque Detection in IVOCT Pullbacks Using Convolutional Neural Networks

Abstract

Coronary heart disease is a common cause of death despite being preventable. To treat the underlying plaque deposits in the arterial walls, intravascular optical coherence tomography can be used by experts to detect and characterize the lesions. In clinical routine, hundreds of images are acquired for each patient which requires automatic plaque detection for fast and accurate decision support. So far, automatic approaches rely on classic machine learning methods and deep learning solutions have rarely been studied. Given the success of deep learning methods with other imaging modalities, a thorough understanding of deep learning-based plaque detection for future clinical decision support systems is required. We address this issue with a new dataset consisting of in-vivo patient images labeled by three trained experts. Using this dataset, we employ state-of-the-art deep learning models that directly learn plaque classification from the images. For improved performance, we study different transfer learning approaches. Furthermore, we investigate the use of cartesian and polar image representations and employ data augmentation techniques tailored to each representation. We fuse both representations in a multi-path architecture for more effective feature exploitation. Last, we address the challenge of plaque differentiation in addition to detection. Overall, we find that our combined model performs best with an accuracy of 91.7%, a sensitivity of 90.9% and a specificity of 92.4%. Our results indicate that building a deep learning-based clinical decision support system for plaque detection is feasible.

Atherosclerotic Plaques Recognition in Intracoronary Optical Images Using Neural Networks

“…When it comes to plaques detection in IOCT, the survey of Boi et al [8] describes that methods using texture information for plaque detection and classification performs with accuracy ranging between 80.41% and 88%. The recent paper of Gessert et al [6] report an accuracy of 92% for plaque detection using texture information. It is interesting to notice that our method, that takes into account only the lumen geometry was able to achieve 91% of accuracy.…”

“…Several papers address atherosclerotic plaques classification in IOCT through the analysis of 2D slices of the vascular wall and propose methods usually very related to texture pattern recognition [3,4]. For instance, Athanasiou et al [5] used Random Forest to analyze textures and classify different plaques and [6] proposed a method to detect and classify plaques between calcified and fibrous/lipid using deep learning model, such as Resnet50 and Densenet-121. Macedo et al [7] proposed to identify side branches in slices using only the lumen contour and reported that irregular geometry in slices with plaque was the main cause of confusion since regular slices without plaques present a nearly circular lumen shape.…”

A novel automated lumen segmentation and classification algorithm for detection of irregular protrusion after stents deployment

“…The features vectors for classification consist of light attenuation coefficients and textures based on gray level co‐occurrence matrix in vessel layer. Gessert et al used convolutional neural networks (CNNs) as a machine learning algorithm to classify plaque using IVOCT images. However, the size of dataset and the number of expert consultants for IVOCT image assessment were small for further plaque detection.…”

Feasibility of Colon Cancer Detection in Confocal Laser Microscopy Images Using Convolution Neural Networks

“…In particular, we investigate both the differentiability of organs and also of malignant and non-malignant tissue both for the colon and peritoneum. As we are dealing with a very small dataset we employ transfer learning which has been shown to improve performance for a variety of medical learning problems [6,7]. We use the state-of-the-art models Densenet121 [8] and SE-Resnext50 [9] which are pretrained on the ImageNet dataset.…”