This image shows how the algorithm first processes a normally stained tissue section (left) and creates a map that allows different tissue types to be displayed in different colors (blue is the tumor, in this case a lung adenocarcinoma). Credit: Dr. Yuri Tolkach
A team of researchers from the Medical Faculty of the University of Cologne and University Hospital Cologne, led by Dr. Juri Tolkach and Prof. Reinhard Büttner, has developed an artificial intelligence-based digital pathology platform that uses new algorithms developed by the team to enable fully automated analysis of tissue sections from patients with lung cancer.
The platform enables computational analysis of digitized tissue samples to detect lung tumors faster and more accurately than ever before. The study, “Next-generation lung cancer pathology: development and validation of diagnostic and prognostic algorithms,” was published in Cell Reports Medicine.
Lung cancer is one of the most common tumors/cancers in humans and has a very high mortality rate. Today, the treatment options for lung cancer patients are determined by pathological examination. Pathologists can also molecularly identify specific genetic alterations to provide personalized treatment.
Over the past few years, pathology has been on the path of digitalization, which has resulted in the obsolete need for microscopes: typical tissue sections are now digitized and analyzed on a computer screen.
Digitalization is essential for the application of advanced analytical methods based on artificial intelligence, which can be used to extract additional cancer information from pathology tissue sections that would not be possible without AI techniques.
“We also show how this platform can be used to develop new clinical tools that can not only improve the quality of diagnostics but also provide new types of information about a patient’s disease, such as how they are responding to treatment,” explained Dr. Yuri Tolkach, physician at the Institute of General Pathology and Pathologic Anatomy at University Hospital Cologne, who led the study.
To prove the broad applicability of the platform, the research team plans to conduct validation studies in collaboration with five pathology laboratories in Germany, Austria and Japan.
Further information: Next-generation lung cancer pathology: development and validation of diagnostic and prognostic algorithms, Cell Reports Medicine (2024). DOI: 10.1016/j.xcrm.2024.101697. www.cell.com/cell-reports-medi … 2666-3791(24)00418-X
Provided by University of Cologne
Source: AI Platform Improves Lung Cancer Diagnosis Accuracy (August 23, 2024) Retrieved August 23, 2024 from https://medicalxpress.com/news/2024-08-ai-platform-lung-cancer-diagnosis.html
This document is subject to copyright. It may not be reproduced without written permission, except for fair dealing for the purposes of personal study or research. The content is provided for informational purposes only.
