Screened Web of Science Corpus on Explainable and Trustworthy AI in Medical Image Analysis: Bibliometric, Scoping and Topic-Modelling Data

This dataset contains the processed bibliographic records used to support the study titled “Mapping the Shift from Performance to Trust in Medical Imaging AI: An Integrated Bibliometric, Scoping and Topic-Modelling Review.” The uploaded CSV file contains 801 records and 72 metadata fields extracted from the Web of Science Core Collection. The available fields include publication type, authors, affiliations, article titles, abstracts, author keywords, Keywords Plus, source titles, publication years, citation counts, cited references, DOI information, Web of Science categories, research areas, funding information and open-access status.

The dataset was prepared for bibliometric mapping, scoping classification, publication and citation trend analysis, contributor and institutional analysis, and latent Dirichlet allocation topic modelling. It covers research relating to artificial intelligence in medical image analysis, including convolutional neural networks, transformer architectures, attention mechanisms, explainable artificial intelligence, medical-image classification, detection, segmentation, diagnosis, clinical decision support, interpretability, fairness and clinical translation.