Figures within biomedical articles present essential evidence of the relevance of a publication in a curation workflow. In particular, visual cues of the image modality or experimental methods can help expert curators identify relevant papers from an increasing number of publications. Automating the identification of these content-bearing images can thus be helpful in computer-assisted curation. However, the paucity of labeled datasets and the specialized training required to label such images hinder the development of such tools. To address this problem, we present the design of ANIMO, a labeling system that integrates extraction and segmentation tools to ease the annotation burden. We first introduce two taxonomies of image modalities and experimental methods, derived in collaboration with curators. On the back-end of the system, we process batches of documents and create a labeling task per document. At the front-end, expert curators can access these tasks through a web interface and access the article of interest. We describe the evaluation of this system by a group of biocurators, and the human factor lessons learned from this interdisciplinary experience.
Motivation: Biocuration is the integration of biological information databases for the enhancement of research. Curation of these databases is challenged by the exponential growth of scientific data and literature. Integration of machine learning and artificial intelligence methods into biocuration workflows may help address this challenge. We report on the discussions, ideas, and recommendations gathered from a workshop “AI and biodata resources: implications for sustainability and best practices in biocuration” at the 18th Annual International Biocuration Conference 2025. Results: Participants agreed that while AI offers transformative potential for efficiency and expanded curatorial capacity, its integration faces substantial hurdles. Key challenges revolve around data and model quality. Reproducibility issues and a lack of open, domain-specific training datasets further compound these problems. Broader concerns include inconsistent data standards, underdeveloped ontologies, unstructured data, legacy systems, and underfunded teams. Despite these issues, several successful AI applications were identified, including tools for literature summarization and workflow assistance. Participants emphasized the need for a refined model of human-AI collaboration requiring clear data provenance and transparency, new skills, and avoiding over-reliance on AI-generated data. The workshop ultimately recommended concerted efforts in infrastructure development, standardization, training, and quality assurance to guide the community toward effective human-AI collaboration that maintains scientific rigor.