{"id":"W4414190431","doi":"10.1101/2025.09.12.25335649","title":"Multimodal Machine Learning for Diagnosis of Multiple Sclerosis Using Optical Coherence Tomography in Pediatric Cases","year":2025,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Optical Coherence Tomography Applications","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Hospital for Sick Children","funders":"","keywords":"Optical coherence tomography; Optic nerve; Deep learning; Multiple sclerosis; Nerve fiber layer; Medical imaging; Convolutional neural network","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000315328,0.0004184025,0.0006620092,0.0008547365,0.00007387765,0.00004359213,0.0005168737,0.0003998601,0.00002734805],"category_scores_gemma":[0.0006199942,0.0004721183,0.0003673468,0.001199116,0.0001173221,0.00007106927,0.0002904238,0.0009091481,0.000002054706],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007331209,"about_ca_system_score_gemma":0.00006840452,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003559845,"about_ca_topic_score_gemma":0.0003190757,"domain_scores_codex":[0.9979059,0.00006352219,0.0007442493,0.0005693797,0.0002464946,0.0004705005],"domain_scores_gemma":[0.9973325,0.001743223,0.0001340303,0.0005051892,0.0001530581,0.0001320519],"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":[0.0000161046,0.0001385399,0.8285669,0.001093655,0.00005193038,0.00000271572,0.00007078349,0.1674633,0.0005088022,0.00008443501,0.000008881521,0.001993953],"study_design_scores_gemma":[0.0008019393,0.00007405998,0.3127798,0.0006070225,0.0003198725,0.000001420242,0.00003260957,0.6762566,0.007964916,0.0003965479,0.0000620196,0.000703152],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9763896,0.001959075,0.01916589,0.00002323475,0.0002528845,0.001358959,0.0003827947,0.0002745464,0.0001930349],"genre_scores_gemma":[0.9550455,0.0006839247,0.04244669,0.000005148259,0.00007675324,0.001622823,0.00006539041,0.00004939885,0.000004417626],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5157871,"threshold_uncertainty_score":0.999773,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04713186697120323,"score_gpt":0.2721856804580601,"score_spread":0.2250538134868569,"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."}}