Dr. Yuri Tolkach and Professor Dr. Reinhard Büttner from the Faculty of Medicine and University Hospital Cologne have led a team of researchers in developing an AI-based digital pathology platform. The platform, powered by new algorithms developed by the team, enables fully automated analysis of tissue sections from lung cancer patients. The aim is to allow faster and more accurate analysis of digitized tissue samples for lung tumors than conventional methods.
The research titled 'Next Generation Lung Cancer Pathology: Development and Validation of Diagnostic and Prognostic Algorithms' has been published in the journal Cell Reports Medicine. Lung cancer is one of the most prevalent human malignancies and carries a high mortality rate. Presently, treatment plans for lung cancer patients are determined by pathological examination. Pathologists also identify specific genetic change to deliver a personalized therapy for patients.
Pathology has significantly transformed over the past years as it transitioned digitally. Microscopes no longer occupy center stage as typical tissue sections are digitized and analysed on computer screens. This digitalization is pivotal for the application of advanced analytical methods like artificial intelligence. Interestingly, AI can extract additional information about the cancer from these pathological tissue sections, something that was impossible without the technology.
Dr. Yuri Tolkach underlines how the platform could be harnessed to develop novel clinical tools. "These tools aren’t restricted to enhancing diagnosis quality; they also provide new types of information about the patient's condition, such as the patient's response to treatment," he explained.
In their mission to expand the applicability of the platform, the research group has a validation study in the pipeline. Five pathological institutes located across Germany, Austria, and Japan will collaborate for this study.
Disclaimer: The above article was written with the assistance of AI. The original sources can be found on ScienceDaily.
