Based on literature reviews and prior experience, numerous inconsistencies have been identified in the reporting of diagnostic imaging studies, especially within dental radiology. Recognising that these same concerns apply broadly across all medical imaging, a new framework is urgently required. Existing tools such as CONSORT AI and SPIRIT AI distinguish themselves by narrow orientations centred on trials or protocols, often lacking the architectural depth to encompass the entire diagnostic workflow. The “Completeness, Learnability, Applicability, Interpretability, Reproducibility, and Evaluation” (CLAIRE) framework aims to bridge this gap. It is presented in three components: (i) a table detailing the importance of each featured concept, (ii) a checklist, and (iii) a grading score to assess the overall reporting quality of a manuscript. CLAIRE seamlessly integrates methodological rigour with real world clinical relevance. It operationalises its six core principles into a unified checklist that uniquely mandates the disclosure of raw results for critical appraisal and heavily emphasises interpretability for clinicians who are not experts in the field. Furthermore, by incorporating an objective scoring system, CLAIRE transcends basic reporting to provide a reproducible assessment tool for journal editors and peer reviewers, effectively bridging the gap between technical artificial intelligence development and safe, transparent clinical application.
Registered: 01 April 2026
Last updated:
Contact: Evando Silva-Filho Email: [email protected]