{"id":"W4410946881","doi":"10.2196/64506","title":"Predicting Early-Onset Colorectal Cancer in Individuals Below Screening Age Using Machine Learning and Real-World Data: Case Control Study","year":2025,"lang":"en","type":"article","venue":"JMIR Cancer","topic":"Colorectal Cancer Screening and Detection","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Logistic regression; Medicine; Random forest; Cohort; Machine learning; Confounding; Colorectal cancer; Artificial intelligence; Propensity score matching; Nomogram; Cancer; Oncology; Internal medicine; Computer science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"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":[0.0008108793,0.0002848872,0.0006342396,0.0005629268,0.0003469501,0.0000996584,0.00012025,0.00009570254,0.00006418712],"category_scores_gemma":[0.0001445298,0.0002790171,0.00005061936,0.001158701,0.0001076929,0.0002372515,0.000260812,0.0008034942,4.535169e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004042441,"about_ca_system_score_gemma":0.0002540967,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.1381443,"about_ca_topic_score_gemma":0.2862148,"domain_scores_codex":[0.9977731,0.0002565815,0.0004683094,0.0007322413,0.0003224439,0.0004473312],"domain_scores_gemma":[0.9990124,0.0002669595,0.0002129846,0.0002832083,0.00008484187,0.0001396813],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.002068767,0.00008941159,0.9580962,0.00007608931,0.0002916513,0.0007615126,0.001195042,0.0004466944,0.0003659304,8.194163e-7,0.00003088006,0.03657699],"study_design_scores_gemma":[0.005598844,0.0008288884,0.9413908,0.0006385028,0.0004203284,0.0001505991,0.0009822639,0.04929541,0.0001509729,0.000003100497,0.0003004758,0.0002398769],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9951989,0.001763738,0.000133131,0.0001859676,0.0002725313,0.001658649,0.0003790988,0.000166679,0.0002412923],"genre_scores_gemma":[0.998261,0.00013491,0.000270826,0.0001256689,0.0002716591,0.0004419262,0.00002173999,0.00004108366,0.0004312061],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1480705,"threshold_uncertainty_score":0.9999662,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04037719020193765,"score_gpt":0.3745165215935605,"score_spread":0.3341393313916229,"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."}}