A Multi-center Study on Intraoperative Glioma Grading via Deep Learning on Cryosection Pathology.

in Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc by Xi Liu, Tianyang Sun, Hong Chen, Shuai Wu, Haixia Cheng, Xiaojia Liu, Qi Lai, Kun Wang, Lin Chen, Junfeng Lu, Jun Zhang, Yaping Zou, Yi Chen, Yingchao Liu, Feng Shi, Lei Jin, Dinggang Shen, Jinsong Wu

TLDR

  • The study developed a deep learning model for intraoperative glioma grading using cryosectioned images, achieving high accuracy and potential for clinical implementation.
  • The model reduced inter-observer variability and enhanced diagnostic consistency among pathologists of varying experience levels.
  • The study demonstrates the potential of artificial intelligence in improving the accuracy and efficiency of surgical pathology diagnostics.

Abstract

Intraoperative glioma grading remains a significant challenge, primarily due to the diminished diagnostic attributable to the suboptimal quality of cryosectioned slides. Precise intraoperative diagnosis is instrumental in guiding surgical strategy to balance the resection extent and the neurological function preservation, thereby optimizing patient prognoses. This study developed a model for intraoperative glioma grading via deep learning on cryosectioned images, termed IGGC. The model was trained and validated on The Cancer Genome Atlas (TCGA) datasets and one cohort (n= 1603, n= 628), and tested on five cohorts (n= 213). The IGGC model achieved an AUC value of 0.99 in differentiating between high grade glioma (HGG) and low grade glioma (LGG), and an AUC value of 0.96 in identifying grade 4 glioma. Integrated into the clinical workflow, the IGGC model assisted pathologists of varying experience levels in reducing inter-observer variability and enhancing diagnostic consistency. This integrated diagnostic model possesses the potential for clinical implementation, offering a time-efficient and highly accurate method for the three-grade classification of adult-type diffuse gliomas based on intraoperative cryosectioned slides.

Overview

  • The study aimed to develop a deep learning model for intraoperative glioma grading using cryosectioned images, termed IGGC.
  • The model was trained and validated on TCGA datasets and one cohort (n= 1603, n= 628) and tested on five cohorts (n= 213).
  • The primary objective of the study was to develop a time-efficient and highly accurate method for the three-grade classification of adult-type diffuse gliomas based on intraoperative cryosectioned slides.

Comparative Analysis & Findings

  • The IGGC model achieved an AUC value of 0.99 in differentiating between high-grade glioma (HGG) and low-grade glioma (LGG).
  • The model achieved an AUC value of 0.96 in identifying grade 4 glioma.
  • Integrated into the clinical workflow, the IGGC model reduced inter-observer variability and enhanced diagnostic consistency among pathologists of varying experience levels.

Implications and Future Directions

  • The IGGC model has the potential for clinical implementation, offering a time-efficient and highly accurate method for intraoperative glioma grading.
  • Future studies can focus on further refining the IGGC model to improve its performance and adapting it for use in other types of tumors.
  • The study's findings highlight the importance of integrating artificial intelligence into surgical pathology workflows to enhance diagnostic accuracy and efficiency.
Read Full Article