Document Actions

July

Picture of the Month - July 2025

Tissue Classification in Feline Cancer Using Raman Spectroscopy and Machine Learning

Accurate assessment of tumor margins remains a critical challenge in cancer surgery. Traditionally, tissue classification relies on histological staining and microscopic analysis — a process that is time-consuming and subject to observer variability. In this interdisciplinary research project, we investigate the potential of Raman spectroscopy combined with machine learning as a rapid, label-free alternative for tissue classification.

The illustration depicts the established workflow: Tissue samples from cats are provided by the team of veterinary pathologist Dr. Manfred Henrich. These samples are analyzed using Raman spectroscopy in the research group of Prof. Peter J. Klar, where characteristic spectral fingerprints are acquired from both healthy and cancerous tissue regions. The collected data is then processed and evaluated in the group of Dr. Janis K. Eckhardt using statistical methods and machine learning algorithms. These models are trained to recognize subtle spectral differences, enabling automated and objective classification of tissue types.

This picture was submitted by Dörte Brandau (group of Dr. Janis K. Eckhardt).

Further insights into the research activities of the ZfM groups can be found in the Gallery.