DPU researchers have developed a new approach for evaluating datasets used in AI-based image analysis.
The NucFuseRank framework provides a standardized foundation for training and benchmarking AI models for nuclei segmentation.
Researchers from Danube Private University (DPU) and the Medical University of Vienna have published a new scientific article in the journal Bioengineering. As part of a research project funded by the WWTF (Vienna Science and Technology Fund), they introduce NucFuseRank, a novel framework for the systematic evaluation and ranking of publicly available datasets used to train and assess AI models for nuclei segmentation.
The study provides a standardized basis for comparing datasets and supports the development of robust AI solutions for biomedical image analysis and digital pathology.
The DPU contributors to the publication are Assoc. Prof. Priv.-Doz. Amirreza Mahbod, PhD, Nima Torbati, PhD, Univ.-Prof. Priv.-Doz. Dr. Ramona Woitek, PhD, and Assist. Prof. Dr. Sepideh Hatamikia.
Zur Publikation: https://doi.org/10.3390/bioengineering13080886
