TY - JOUR T1 - Overview of the ImageCLEFmedical 2026 GANs task: Exploring fingerprint detection, privacy leakage, and synthetic medical image generation A1 - Andrei, Alexandra-Georgiana A1 - Constantin, Mihai Gabriel A1 - Dogariu, Mihai A1 - Karpenka, Dzmitry A1 - Prokopchuk, Yuri A1 - Radzhabov, Ahmedkhan A1 - Stanciu, Dan-Cristian A1 - Kovalev, Vassili A1 - Müller, Henning A1 - Ionescu, Bogdan JA - Springer Nature Y1 - 2026 VL - 17087 SP - 465 EP - 492 KW - Deep Learning KW - Generative Adversarial Networks KW - generative models KW - ImageCLEF benchmarking lab KW - latent space KW - medical imaging KW - medical synthetic data KW - Privacy N2 - This paper presents an overview of the ImageCLEF 2026 lab, organized within the Conference and Labs of the Evaluation Forum – CLEF Labs 2026 [17]. This year’s 24th edition of the evaluation event consists of five main tasks: (i) the Medical task, comprised of four sub-tasks, consisting of various challenges in the medical field, like Concept Detection and Caption Prediction on real and synthetic medical images, evaluation of Veracity and Privacy of GAN-generated 3D CT scans, Visual Question Answering and explainability in gastrointestinal endoscopy images, and a new MEDIQA-CORE task on Brain Tumor Subtype Classification and Report Discrepancy Summarization on real-life medical data, (ii) the ToPicto task, focusing on aiding communication using text-to-pictogram translation and pictogram sequence prediction, (iii) a Multimodal Reasoning task on multi-lingual, multi-discipline question answering on multiple modalities, and two new tasks on (iv) Deepfake Detection and Generation in images and audio and (v) the AI4Agriculture task, focused on Viticultultire Potential prediction and Crop Identification. This edition of ImageCLEF attracted a total of 307 teams registered for all the different tasks. This resulted in 1,284 total runs across all competitions, and a total of 74 working notes papers submitted, making this year’s one of the most successful editions. ER -