{"id":"W4407698494","doi":"10.1016/j.eclinm.2025.103089","title":"Use of artificial intelligence with retinal imaging in screening for diabetes-associated complications: systematic review","year":2025,"lang":"en","type":"review","venue":"EClinicalMedicine","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Medical Research Council","keywords":"Medicine; Diabetes mellitus; Retinal; Optometry; Intensive care medicine; Ophthalmology; Endocrinology","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":["metaresearch","metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.003104735,0.0004410415,0.006650691,0.0005541415,0.00005489787,0.0000192102,0.0002966913,0.0001454553,0.00001783143],"category_scores_gemma":[0.01893697,0.0002838806,0.0008489261,0.001651258,0.0002612852,0.00005610082,0.00006586122,0.0006615666,0.000005493871],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001020592,"about_ca_system_score_gemma":0.0003701972,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002378469,"about_ca_topic_score_gemma":0.000006745844,"domain_scores_codex":[0.9943485,0.0004814401,0.003741383,0.000619705,0.0004465644,0.0003623619],"domain_scores_gemma":[0.9899812,0.006628355,0.001854626,0.0009106811,0.0004870375,0.0001381408],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"systematic_review","study_design_scores_codex":[0.00001965373,0.0001888138,0.0004754287,0.8290069,0.0003963032,0.00001535755,0.000007501404,8.209911e-7,2.259629e-7,0.0000360454,0.0003547186,0.1694982],"study_design_scores_gemma":[0.0001783979,0.0002033606,0.00003797587,0.971822,0.02021518,0.0000178672,0.00002248118,0.001311165,4.340914e-7,0.00003424366,0.005938891,0.0002179796],"study_design_candidate":"systematic_review","study_design_consensus":"systematic_review","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.000002959791,0.9864427,0.007471852,0.001910974,0.00004735749,0.003969094,0.00004774383,0.00004701221,0.00006030113],"genre_scores_gemma":[0.00004894828,0.9921641,0.004860199,0.00131481,0.00007922915,0.0006028557,0.0006357328,0.00004563765,0.0002484593],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.1692802,"threshold_uncertainty_score":0.9999613,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1936104279574102,"score_gpt":0.4569677345525127,"score_spread":0.2633573065951025,"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."}}