The SEARCH project has created its own page on Hugging face, a leading open-source platform where the machine learning community can collaborate on models, datasets, and applications. The page currently features two resources developed within the project.

The first resource is a novel variational autoencoder-based model, named TIDE-II, for generating high-resolution synthetic Wireless Capsule Endoscopy (WCE) images. The model can generate clinically plausible, very realistic WCE images, of improved quality compared to relevant state-of-the-art generative models. TIDE-II was implemented in the study ‘Clinical Evaluation of Medical Image Synthesis: A Case Study in Wireless Capsule Endoscopy’, led by Dr. Iakovidis Dimitrios, Director of the Biomedical Imaging Laboratory at the Department of Computer Science and Biomedical Informatics, University of Thessaly, Greece.

The second tool allows the creation of realistic synthetic electrocardiogram (ECG) signals that can be used for research, algorithm testing, and AI training without exposing sensitive patient information. The resource, named Deepfake ECG Generator – Plus, has been shared by Vajira Thambawita, research scientist at Simula, Norway.

SEARCH partners will regularly update the platform with additional resources, models, and demonstrations as the project progresses.