{"id":"W4391264227","doi":"10.1016/j.oret.2024.01.019","title":"Deep Learning Models for the Screening of Cognitive Impairment Using Multimodal Fundus Images","year":2024,"lang":"en","type":"article","venue":"Ophthalmology Retina","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Beijing Municipal Science and Technology Commission; Capital Health Research and Development of Special Fund; National Natural Science Foundation of China","keywords":"Fundus (uterus); Beijing; Medicine; Optical coherence tomography; Optometry; Montreal Cognitive Assessment; Ophthalmology; Cognitive impairment; Artificial intelligence; Cognition; Computer science; China","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":[],"consensus_categories":[],"category_scores_codex":[0.0005388081,0.0001628719,0.0003503704,0.0001381721,0.0001760954,0.0000288466,0.00008347107,0.00008820468,0.00009658932],"category_scores_gemma":[0.0002625696,0.0001106653,0.0002846013,0.0002367178,0.0002462398,0.00007378704,0.00005965052,0.0003414279,0.000005204953],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002948265,"about_ca_system_score_gemma":0.00004790883,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00014278,"about_ca_topic_score_gemma":1.612081e-7,"domain_scores_codex":[0.9988471,0.0001000288,0.0002728518,0.0003216702,0.0001596178,0.000298775],"domain_scores_gemma":[0.9985409,0.0009725147,0.00009375387,0.0001438263,0.0001867853,0.00006226773],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.008570581,0.001624553,0.4359652,0.005077734,0.01252074,0.01033924,0.01504251,0.1906683,0.0926882,0.001309505,0.000833346,0.22536],"study_design_scores_gemma":[0.0006458894,0.0005759237,0.005394061,0.0007689882,0.001185298,0.003672904,0.002101365,0.9832425,0.002007273,0.0002360737,0.00003361178,0.0001361163],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8505906,0.005138884,0.1406773,0.0009032851,0.0001213634,0.0004554971,0.00001097269,0.00006362606,0.002038489],"genre_scores_gemma":[0.9903975,0.00004075547,0.007936526,0.00002918374,0.0001407193,0.00003390891,0.00002646228,0.00003411746,0.001360857],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7925742,"threshold_uncertainty_score":0.4512799,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06022062300058528,"score_gpt":0.3573060337078574,"score_spread":0.2970854107072721,"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."}}