{"id":"W4378470023","doi":"10.3389/fneur.2023.1165267","title":"Accounting for uncertainty in training data to improve machine learning performance in predicting new disease activity in early multiple sclerosis","year":2023,"lang":"en","type":"article","venue":"Frontiers in Neurology","topic":"Multiple Sclerosis Research Studies","field":"Medicine","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary; University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Random forest; Probabilistic logic; Missing data; Artificial intelligence; Medicine; Clinical trial; Machine learning; Statistics; Computer science; Internal medicine; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02415439,0.001959275,0.001500571,0.001582173,0.0007489945,0.00142058,0.001162926,0.002032604,0.00099074],"category_scores_gemma":[0.03338592,0.0006102514,0.001951905,0.0009324143,0.0007357831,0.00195986,0.001198318,0.002145698,0.0004241701],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001101844,"about_ca_system_score_gemma":0.001441927,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007745987,"about_ca_topic_score_gemma":0.005914088,"domain_scores_codex":[0.9939706,0.003776133,0.0003941373,0.001094842,0.0004422163,0.0003220633],"domain_scores_gemma":[0.9683174,0.02632223,0.001877432,0.001411826,0.001725968,0.0003452286],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001370578,0.0004630346,0.07478685,0.0002872745,0.001082867,0.0001182661,0.000166171,0.7741861,0.00219883,0.0006741644,0.001866052,0.1427998],"study_design_scores_gemma":[0.00006609331,0.0006327241,0.01148366,0.0001167731,0.0001620874,0.00008440324,0.00003259334,0.9818805,0.00210559,0.002829982,0.0005496109,0.00005602049],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6613707,0.007075017,0.3236459,0.00169053,0.0002286754,0.0002512661,0.001111575,0.002620774,0.002005537],"genre_scores_gemma":[0.9543978,0.0003585007,0.04357629,0.0002807614,0.00008406062,0.00009456022,0.0008318257,0.00005794558,0.0003182047],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02415439,"threshold_uncertainty_score":0.1277422,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1194036308837635,"score_gpt":0.3176845510673278,"score_spread":0.1982809201835643,"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."}}