Segmentation and Classification in Digital Pathology for Glioma Research: Challenges and Deep Learning Approaches

TLDR

Biomedical imaging supplies essential data for cancer research, enabling the characterization of tumor morphology across disease stages and treatment responses, yet extracting and classifying visual and latent features remains increasingly difficult due to the growing complexity and resolution of biomedical image data. The study introduces four deep‑learning image analysis methods developed at the CPM satellite event of MICCAI 2018 to advance glioma research. The authors built four deep‑learning models: a nuclei segmentation network for whole‑slide images and three classification networks that fuse radiographic and histologic image data to differentiate oligodendroglioma from astrocytoma. The models achieved a Dice similarity of 0.868 for nuclei segmentation and classification accuracies of 0.75, 0.80, and 0.90, showing that carefully designed deep‑learning algorithms can reach high performance and that combining radiographic with histologic information improves classification.

Abstract

Biomedical imaging Is an important source of information in cancer research. Characterizations of cancer morphology at onset, progression, and in response to treatment provide complementary information to that gleaned from genomics and clinical data. Accurate extraction and classification of both visual and latent image features Is an increasingly complex challenge due to the increased complexity and resolution of biomedical image data. In this paper, we present four deep learning-based image analysis methods from the Computational Precision Medicine (CPM) satellite event of the 21st International Medical Image Computing and Computer Assisted Intervention (MICCAI 2018) conference. One method Is a segmentation method designed to segment nuclei in whole slide tissue images (WSIs) of adult diffuse glioma cases. It achieved a Dice similarity coefficient of 0.868 with the CPM challenge datasets. Three methods are classification methods developed to categorize adult diffuse glioma cases into oligodendroglioma and astrocytoma classes using radiographic and histologic image data. These methods achieved accuracy values of 0.75, 0.80, and 0.90, measured as the ratio of the number of correct classifications to the number of total cases, with the challenge datasets. The evaluations of the four methods indicate that (1) carefully constructed deep learning algorithms are able to produce high accuracy in the analysis of biomedical image data and (2) the combination of radiographic with histologic image information improves classification performance.