Clinical Cyberbioethics and AI-Mediated Clinical Decision-Making: a mapping and narrative review with thematic synthesis (SWiM)

Background: Artificial intelligence (AI) is rapidly reshaping clinical decision-making, yet it introduces ethically salient risks including bias, automation bias, opacity, privacy threats, data drift, adversarial vulnerability, and unclear accountability. Objective: To define and operationalize Clinical Cyberbioethics as an integrative framework for AI-mediated clinical decision-making, synthesizing ethical principles, governance requirements, and patient-facing digital rights proposed across the literature. Methods: We conducted a mapping and narrative review with SWiM-guided thematic synthesis covering January 2015 to March 2025, and documented identification, screening, and inclusion decisions using a PRISMA 2020–adapted flow diagram. Records were identified from an author-curated reference set compiled during the stated search period; screening and selection are reported in Fig. 1 and Supplementary Files 1–4. Results: Records identified (n = 54); duplicates removed (n = 11); records screened (n = 43); records excluded at title/abstract (n = 1); full-text sources assessed for eligibility (n = 42); studies included in synthesis (n = 42); full-text exclusions with reasons (n = 0). The synthesis yielded operational domains spanning decision authority, explainability/contestability, bias and equity safeguards, lifecycle governance and accountability, and proportional data governance and privacy safeguards. Conclusions: Clinical Cyberbioethics provides a practical framework to guide trustworthy AI-mediated clinical decision-making by integrating ethical principles with governance and patient rights.