AI and nanopore sequencing to enable faster brain tumor diagnosis

The international team, including researchers from Oslo University Hospital (Norway), Acibadem University (Turkey), and the University Health Network in Toronto (Canada), aims to enhance diagnostic procedures through a combination of artificial intelligence (AI) and new laboratory technology. These methods will allow for the rapid and precise classification of tumor cells, making it easier to determine which therapies will be most effective and which the tumor may be resistant to. 

“The software is already freely available worldwide for scientific purposes, but the collection of data to be fed into the software is still lengthy. With DC2M-TAEC, we will develop a faster and more accessible method,” says Professor Sahm, Deputy Medical Director of the Department of Neuropathology at the Institute of Pathology at the Heidelberg University Hospital. 

The research team is building on previous work conducted at Heidelberg University Hospital, the Heidelberg Medical Faculty, the Hopp Children's Tumor Centre Heidelberg, and the German Cancer Research Center. Their existing AI-based molecular diagnostic method analyzes tumor DNA methylation patterns to classify brain tumors. However, current data collection remains time-consuming. The new project seeks to optimize this by integrating AI-driven classification with nanopore sequencing, which detects DNA sequences based on electrical signals. 

By refining this approach, the researchers aim to make rapid tumor classification available during surgical procedures. “The aim is to make the analysis results from the tissue samples available very quickly, if possible during the operation itself,” explains Prof. Sahm. This advancement will also help set up cost-effective diagnostic laboratories in various locations worldwide, particularly in low- and middle-income countries. 

Each participating institution has a specific focus in the project: 

• The Heidelberg team is refining bioinformatics methods for methylation-based classification and adapting the algorithm to process various types of sequencing data. 

• Researchers in Oslo are optimizing sample preparation and sequencing techniques for real-time tumor classification during surgery. 

• The team in Istanbul is evaluating the feasibility of implementing the procedure in healthcare systems with limited resources. 

• Researchers in Toronto are advancing nanopore sequencing for liquid biopsies, such as blood and spinal fluid, to facilitate tumor identification before surgery. 

MEDICA-tradefair.com; Source: Heidelberg University Hospital