{"id":"W4386470319","doi":"10.1109/tmi.2023.3312524","title":"Clinically-Inspired Multi-Agent Transformers for Disease Trajectory Forecasting From Multimodal Data","year":2023,"lang":"en","type":"article","venue":"IEEE Transactions on Medical Imaging","topic":"Osteoarthritis Treatment and Mechanisms","field":"Medicine","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute on Aging; Infotech Oulu; Canadian Institutes of Health Research; National Institutes of Health; CSC – IT Center for Science; IXICO; H. Lundbeck A/S; Genentech; Vlaamse regering; Servier; Eli Lilly and Company; U.S. Department of Defense; China Scholarship Council; Eisai; Alzheimer's Disease Neuroimaging Initiative; Academy of Finland; Northern California Institute for Research and Education; Alfred Kordelinin Säätiö; Pfizer; Biogen; BioClinica; University of Southern California; GlaxoSmithKline; Oulun Yliopisto; Meso Scale Diagnostics; Novartis Pharmaceuticals Corporation; Bristol-Myers Squibb; Alzheimer's Association; U.S. Department of Health and Human Services","keywords":"Computer science; Artificial intelligence; Transformer; Machine learning; Trajectory; Deep learning; Data mining","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005462779,0.0002827715,0.0004249362,0.0002005397,0.000289435,0.0000301783,0.0002371028,0.0001296483,0.0003704061],"category_scores_gemma":[0.0001770922,0.000255012,0.0003154835,0.0002462865,0.0001470453,0.000221984,0.000004368385,0.0004394967,0.0001141009],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007804408,"about_ca_system_score_gemma":0.0003147836,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007383896,"about_ca_topic_score_gemma":0.00009106299,"domain_scores_codex":[0.99732,0.00005948642,0.0006029006,0.0007496894,0.00070441,0.0005635467],"domain_scores_gemma":[0.9977958,0.0006133083,0.00006554716,0.0004725733,0.00006230859,0.0009904662],"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":[0.0009797543,0.0007930869,0.000490164,0.0000906276,0.0001417043,0.0005405306,0.0002659912,0.00005255425,0.002432718,0.000001055977,0.0003948162,0.993817],"study_design_scores_gemma":[0.02118676,0.0005637643,0.0007718724,0.0009377752,0.001049128,0.00002845701,0.0007563335,0.9679011,0.003657871,0.00004458902,0.002642782,0.0004595797],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1092968,0.0001498104,0.8821495,0.003262758,0.001875886,0.001309747,0.001278028,0.0006159943,0.00006150224],"genre_scores_gemma":[0.9870813,0.0002012252,0.009460867,0.001191308,0.0003317159,0.0002679644,0.00100313,0.00008732104,0.000375156],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9933574,"threshold_uncertainty_score":0.9999902,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0925703026526429,"score_gpt":0.3525508274316693,"score_spread":0.2599805247790264,"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."}}