Medical Report Generation
Overview
Medical report generation studies how to create diagnostic reports and clinical notes from images, patient records, or doctor–patient conversations. Approaches include vision-language models for radiology, pathology, and other imaging; speech-to-text pipelines for SOAP notes; and systems that use templates, retrieval, structured findings, or evidence-linked reasoning to guide generation. Evaluation increasingly uses clinical measures and expert review alongside language metrics on settings such as MIMIC-CXR and IU X-Ray. A central challenge is ensuring that reports are complete and factually grounded across hospitals and modalities, since fluent text and lexical similarity do not guarantee clinical accuracy or safety.
Latest papers
82 papers12 papers in the last four weeks, up 33% on the four weeks before. 0.1% of all new papers.