{"id":"W4415981229","doi":"10.48550/arxiv.2510.02325","title":"Agentic-AI Healthcare: Multilingual, Privacy-First Framework with MCP Agents","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Multi-Agent Systems and Negotiation","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Audit; Certification; Context (archaeology); Protocol (science); Health care; Architecture; Data access; Access control","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005270959,0.0006277273,0.0007125977,0.0002975182,0.0004355659,0.00035949,0.002462561,0.0006704115,0.00004406252],"category_scores_gemma":[0.0002332891,0.0005648702,0.000245548,0.0005494835,0.00006183024,0.0003192039,0.002444071,0.001376037,0.0002651752],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00027553,"about_ca_system_score_gemma":0.0005426425,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00122998,"about_ca_topic_score_gemma":0.0002604447,"domain_scores_codex":[0.9957587,0.0002466315,0.0008534479,0.00167302,0.0007611651,0.0007070881],"domain_scores_gemma":[0.995621,0.000198385,0.0006533255,0.002886533,0.000348958,0.0002918605],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00006080037,0.0005866471,0.9630617,0.002867382,0.0004935314,0.0002030492,0.01107961,0.002418669,0.00004791894,0.004539504,0.008867731,0.005773446],"study_design_scores_gemma":[0.002252817,0.0002596766,0.8422774,0.007765726,0.0001699298,0.00002466606,0.0001139589,0.06376569,0.002291144,0.002202673,0.0765445,0.002331824],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5965469,0.0009207659,0.3820399,0.01100294,0.006416488,0.001940111,0.00004195054,0.0007274114,0.0003635058],"genre_scores_gemma":[0.9735828,0.000229749,0.02095158,0.0023369,0.0005046753,0.0001759927,0.00005597286,0.00004307024,0.00211925],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3770359,"threshold_uncertainty_score":0.9996803,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07230628956691797,"score_gpt":0.3356078838759599,"score_spread":0.2633015943090419,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}