{"id":"W4409640249","doi":"10.1101/2025.04.11.648433","title":"Deep learning using structural MRI massively improves prediction accuracy of body mass index","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Body Composition Measurement Techniques","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University; St. Joseph’s Healthcare Hamilton","funders":"","keywords":"Index (typography); Deep learning; Artificial intelligence; Body mass index; Computer science; Massively parallel; Pattern recognition (psychology); Medicine; Internal medicine; Parallel computing","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001134331,0.0007655472,0.0005325434,0.000657118,0.0001393553,0.0006300026,0.0005768679,0.0006143001,0.002713797],"category_scores_gemma":[0.003067356,0.0003384334,0.0004845661,0.0003646226,0.0002833831,0.000548429,0.0006572956,0.0009368784,0.0008449439],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003314618,"about_ca_system_score_gemma":0.0006477139,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005622835,"about_ca_topic_score_gemma":0.006618395,"domain_scores_codex":[0.999836,0.00005128829,0.000008358,0.00005342358,0.00002596063,0.00002497095],"domain_scores_gemma":[0.9991853,0.0003499799,0.0000944803,0.0001367347,0.0001512496,0.00008233004],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001833502,0.000662424,0.1051203,0.0002587754,0.0006553071,0.0002951179,0.00008285216,0.3039579,0.04254126,0.00312585,0.02466707,0.5167997],"study_design_scores_gemma":[0.00003216758,0.00009090185,0.01005266,0.00002346644,0.00004642555,0.00005893845,0.00001068287,0.9792188,0.00523536,0.004385369,0.000828513,0.00001677554],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7496584,0.00229429,0.2312345,0.004201702,0.0003589513,0.00007515151,0.002818077,0.004472331,0.004886627],"genre_scores_gemma":[0.9536289,0.0004383409,0.0409437,0.0004025155,0.0001188232,0.0000305465,0.001656223,0.0001155124,0.002665448],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005622835,"threshold_uncertainty_score":0.01118022,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01611878280925033,"score_gpt":0.2600632205906222,"score_spread":0.2439444377813719,"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."}}