TY  - JOUR
T1  - Overview of ImageCLEFmedical 2025 – Medical Concept Detection and Interpretable Caption Generation
A1  - Damm, Hendrik
A1  - Pakull, Tabea M. G.
A1  - Becker, Helmut
A1  - Bracke, Benjamin
A1  - Eryilmaz, Bahadir
A1  - Bloch, Louise
A1  - Brüngel, Raphael
A1  - Schmidt, Cynthia S.
A1  - Rückert, Johannes
A1  - Pelka, Obioma
A1  - Schäfer, Henning
A1  - Idrissi-Yaghir, Ahmad
A1  - Abacha, Asma Ben
A1  - de Herrera, Alba G. Seco
A1  - Müller, Henning
A1  - Friedrich, Christoph M.
JA  - CEUR-WS.org/Vol-4038 - CLEF 2025 Working Notes
Y1  - 2025
VL  - 4038
IS  - Paper 170
UR  - https://ceur-ws.org/Vol-4038/paper_170.pdf
KW  - computer vision
KW  - Explainable AI
KW  - Image Captioning
KW  - Image Understanding
KW  - ImageCLEF
KW  - Multi-label classification
KW  - Radiology
N2  - The ImageCLEFmedical 2025 Caption task follows challenges held from 2017–2024 and comprises three subtasks:
concept detection, caption prediction, and a newly introduced explainability task. The goal is to extract Unified
Medical Language System (UMLS) concepts, generate fluent captions from medical images, and provide humaninterpretable justifications for the outputs. This year’s edition used an enlarged version of the Radiology Objects
in COntext version 2 (ROCOv2) dataset, which was expanded with new articles and the inclusion of the optical
coherence tomography (OCT) imaging modality. For concept detection, the F1-score was used to evaluate
predictions against UMLS terms. For caption prediction, evaluation was updated to a composite score averaging
six metrics to assess both relevance and factuality. The new explainability submissions were manually judged by a
radiologist. The 2025 task attracted 80 registered research groups, with 11 teams submitting a total of 149 graded
runs across the three subtasks. Top-performing systems for concept detection were predominantly based on
ensembles of Convolutional Neural Networks (CNNs). For caption prediction, a general shift towards fine-tuning
Vision-Language Models (VLMs) was observed, with adapted architectures like BLIP leading to strong results
across the new composite metrics. Finally, the inaugural explainability task saw initial submissions of post-hoc
visualizations, establishing a baseline and clarifying the need for model-intrinsic explanations in future editions.
ER  -