{"id":"W2964363195","doi":"10.1002/hbm.24682","title":"Modeling longitudinal imaging biomarkers with parametric Bayesian multi‐task learning","year":2019,"lang":"en","type":"article","venue":"Human Brain Mapping","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"FP7 Information and Communication Technologies; National Institute on Aging; Horizon 2020 Framework Programme; Medical Research Council Canada; Nederlandse Organisatie voor Wetenschappelijk Onderzoek; Medical Research Council; Engineering and Physical Sciences Research Council; National Institute for Health and Care Research","keywords":"Computer science; Artificial intelligence; Machine learning; Neuroimaging; Bayesian probability; Biomarker; Parametric statistics; Alzheimer's Disease Neuroimaging Initiative; Multi-task learning; Metric (unit); Disease; Task (project management); Alzheimer's disease; Medicine; Psychology; Neuroscience; Statistics; Mathematics; Pathology; Biology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007985656,0.0002302625,0.0003057453,0.0007510829,0.0003467711,0.000125291,0.0001263584,0.0000474614,0.0005587663],"category_scores_gemma":[0.0001423058,0.0002013753,0.0001075608,0.0006917305,0.00007723573,0.000206099,0.00009694847,0.0004638476,0.0001537659],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001136415,"about_ca_system_score_gemma":0.00007480966,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000996602,"about_ca_topic_score_gemma":0.000008347297,"domain_scores_codex":[0.9978489,0.0001120676,0.000280703,0.0005529195,0.0005582455,0.0006471163],"domain_scores_gemma":[0.9991953,0.0001053701,0.00007176613,0.0002597538,0.0001715039,0.0001962508],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00009499575,0.000115564,0.9600309,0.0002089939,0.0002054714,0.000160524,0.0002760339,0.0007417334,0.0344679,0.00008070272,0.00008477693,0.003532473],"study_design_scores_gemma":[0.003949586,0.0003322739,0.5960181,0.0006593613,0.00006323709,0.0000999836,0.002319066,0.3954434,0.00008500629,0.00004777954,0.0006498866,0.0003323735],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8866079,0.0001529533,0.1042896,0.0006751384,0.00003575281,0.0006596672,6.756585e-7,0.0001455813,0.007432735],"genre_scores_gemma":[0.9941449,0.000005119834,0.002354715,0.0003054871,0.00008080865,0.00002893601,0.00004155271,0.00004928457,0.002989221],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3947016,"threshold_uncertainty_score":0.8211847,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03068147025008076,"score_gpt":0.3096311163896315,"score_spread":0.2789496461395507,"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."}}