Generating 2.5D pathology for enhanced viewing and AI diagnosis

TL;DR

This paper introduces a new method to create 2.5D pathology images from biopsy samples, improving both AI diagnosis and pathologist visualization.

Contribution

A novel morphology-preserving alignment framework to generate 2.5D biopsy cores for enhanced AI and pathologist use.

Findings

The framework was applied to over 12,000 biopsy samples across prostate, breast, and renal tissues.

A deep learning model trained on 2.5D cores showed improved cancer grading performance.

Pathologists confirmed the 2.5D cores improved their ability to evaluate tissue structures.

Abstract

Histological analysis of biopsy samples by pathologists can require the evaluation of complex three-dimensional (3D) tissue structures. This process involves studying the same tissue region across slides, which requires laborious zooming and panning for localization. Additionally, standard deep learning frameworks typically focus on cross-sections cut from biopsy specimens, limiting their ability to capture 3D tissue spatial information. We present a novel framework that constructs 2.5D biopsy cores via the extraction and co-alignment of serial tissue sections using a novel morphology-preserving alignment framework. These 2.5D cores can then be used for enhanced viewing by pathologists and as input to video transformer models that can capture depth-wide spatial dependencies. We used our framework to construct 2.5D cores for 10,210 prostate biopsies, 156 breast biopsies, and 1869 renal biopsies. To evaluate the utility of the cores for downstream tasks, we performed additional studies in prostate cancer by: (1) training a deep learning-based cancer grading model and (2) conducting a reader study with pathologists.

Figures