We use cookies to understand how you use our site and to improve your experience. This includes personalizing content and advertising. To learn more, click here. By continuing to use our site, you accept our use of cookies. Cookie Policy.

Features Partner Sites Information LinkXpress
Sign In
Advertise with Us
LGC Clinical Diagnostics

Download Mobile App

AI Model for Brain Tumor Classification Advances Neuropathology

By LabMedica International staff writers
Posted on 28 Dec 2023
Print article
Image: Ten examples of classification results on the external testing sets (Photo courtesy of Nature Communications, 2023)
Image: Ten examples of classification results on the external testing sets (Photo courtesy of Nature Communications, 2023)

Diffuse gliomas, which comprise a large portion of malignant brain tumors in adults, include various types such as astrocytoma, oligodendroglioma, and glioblastoma. Diagnosing these types of gliomas traditionally relies on an analysis that integrates histological characteristics with molecular details, a method that presents significant complexities when attempting to develop a comprehensive diagnostic model from whole-slide images (WSIs). The immense gigapixel resolution of WSIs renders the use of standard convolutional neural networks for analysis impractical. To address this challenge, researchers have now introduced a novel integrated diagnostic model that can automatically classify adult-type diffuse gliomas directly from unannotated standard whole-slide pathological images, eliminating the need for additional molecular testing.

Researchers from the Chinese Academy of Sciences (CAS, Beijing, China) have devised this deep learning model capable of parsing WSIs and categorizing gliomas without the need for detailed manual annotations. This model adheres to the strict classification guidelines outlined in the 2021 fifth edition of the World Health Organization Classification of Tumors of the Central Nervous System. The model underwent training and validation across a diverse dataset comprising 2,624 patient cases collected from three different hospitals.

The model's effectiveness was evaluated based on its classification accuracy, sensitivity to various glioma types and grades, and its capability to differentiate between genotypes that exhibit similar histological characteristics. The outcomes of the experiments indicate that the model demonstrates robust performance, with all areas under the receiver operator curve exceeding 0.90. This performance was noted in its ability to classify major tumor types, identify tumor grades within each type, and, notably, distinguish between tumor genotypes that share the same histological features.

"Our integrated diagnosis model has the potential to be used in clinical scenarios for automated and unbiased classification of adult-type diffuse gliomas," said CAS Prof. Li Zhicheng who led the research team. "The future research will focus on improving this model to have multi-center, multi-racial datasets."

Related Links:
Chinese Academy of Sciences

Platinum Member
COVID-19 Rapid Test
OSOM COVID-19 Antigen Rapid Test
Specimen Collection & Transport
Anti-Cyclic Citrullinated Peptide Test
GPP-100 Anti-CCP Kit
Gold Member
Turbidimetric Control
D-Dimer Turbidimetric Control

Print article


Molecular Diagnostics

view channel
Image: Prostate cancer cell image taken using a scanning electron microscope (Photo courtesy of LRI EM Unit)

New Discoveries of Prostate Cancer Evolution Pave Way for Genetic Test

Prostate cancer ranks as one of the most common cancers affecting men, and while it accounts for a significant number of male cancer fatalities, many men live with it rather than die from it.... Read more


view channel
Image: The Gazelle Hb Variant Test (Photo courtesy of Hemex Health)

First Affordable and Rapid Test for Beta Thalassemia Demonstrates 99% Diagnostic Accuracy

Hemoglobin disorders rank as some of the most prevalent monogenic diseases globally. Among various hemoglobin disorders, beta thalassemia, a hereditary blood disorder, affects about 1.5% of the world's... Read more


view channel
Image: The companies will develop genetic testing systems based on capillary electrophoresis sequencers (Photo courtesy of 123RF)

Sysmex and Hitachi Collaborate on Development of New Genetic Testing Systems

Sysmex Corporation (Kobe, Japan) and Hitachi High-Tech Corporation (Tokyo, Japan) have entered into a collaboration for the development of genetic testing systems using capillary electrophoresis sequencers... Read more
Copyright © 2000-2024 Globetech Media. All rights reserved.