HistoSegCap: Capsules for Weakly-Supervised Semantic Segmentation of Histological Tissue Type in Whole Slide Images

TL;DR

Experimental results show that the proposed model outperforms traditional methods in terms of accuracy and the mean Intersection-over-Union (mIoU) metric, and underscore the potential of Capsule Networks in enhancing the precision and efficiency of histopathological image analysis.

Abstract

Digital pathology involves converting physical tissue slides into high-resolution Whole Slide Images (WSIs), which pathologists analyze for disease-affected tissues. However, large histology slides with numerous microscopic fields pose challenges for visual search. To aid pathologists, Computer Aided Diagnosis (CAD) systems offer visual assistance in efficiently examining WSIs and identifying diagnostically relevant regions. This paper presents a novel histopathological image analysis method employing Weakly Supervised Semantic Segmentation (WSSS) based on Capsule Networks, the first such application. The proposed model is evaluated using the Atlas of Digital Pathology (ADP) dataset and its performance is compared with other histopathological semantic segmentation methodologies. The findings underscore the potential of Capsule Networks in enhancing the precision and efficiency of histopathological image analysis. Experimental results show that the proposed model outperforms traditional methods in terms of accuracy and the mean Intersection-over-Union (mIoU) metric.

Figures (11)

• Figure 1: The proposed methodology utilizes training based on annotations of histological tissue segments and forecasts both the morphological and functional types of tissue at the granular level of individual pixels.
• Figure 2: The proposed Atlas database employs a structured classification of histological tissue types for guided annotation. This tissue classification system is organized into three tiers, progressing from the broadest category at the top to the most detailed at the bottom.
• Figure 3: Proposed HistoSegCap architecture.
• Figure 4: Proposed reconstruction architecture.
• Figure 5: This figure demonstrates the application of image reconstruction technique in generating output results for a specific tissue type. The process enhances the clarity and detail of the tissue image, thereby improving the accuracy of the results.