Maiju Lepomäki successfully defended her PhD dissertation on breast surgery margin assessment and complications in April, congratulations Maiju! The work consisted of four original peer-reviewed publications covering the topic from many viewpoints. Two of the publications considered the use of DMS in margin assessment with automated tissue analysis systems.

In study II, the team studied benign and malignant punch biopsies from breast cancer patients. Sampling was performed with an automated tissue analysis system (ATAS), which utilized monopolar diathermy for sample evaporation.

In study IV, different machine learning methods were put to the test with a large animal model consisting of five different tissue types. Nonlinear classification methods were found to outperform linear ones, with CNN working especially well. The same sampling method, based on a laser cutter, was used to map cross-sections of three breast carcinomas to test the feasibility of the system for pathological use.

The overall finding was that evaporation-based sampling and DMS analysis is suitable for tissue identification. In the future, DMS analysis could be efficient both in tissue imagining before traditional microscopy, and intraoperative margin assessment.

The full dissertation is available on the Tampere University library site (in English) here.