{"id":"W4407668521","doi":"10.3389/fninf.2025.1527582","title":"Contrastive self-supervised learning for neurodegenerative disorder classification","year":2025,"lang":"en","type":"article","venue":"Frontiers in Neuroinformatics","topic":"Dementia and Cognitive Impairment Research","field":"Medicine","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institutes of Health Research; University of California, San Francisco; National Institutes of Health; Genentech; IXICO; H. Lundbeck A/S; Servier; Eisai; Deutsche Forschungsgemeinschaft; Northern California Institute for Research and Education; Pfizer; Novartis Pharmaceuticals Corporation; University of Southern California; Biogen; Eli Lilly and Company; Bristol-Myers Squibb; BioClinica; U.S. Department of Defense; Alzheimer's Disease Neuroimaging Initiative; Meso Scale Diagnostics; National Institute on Aging; Alzheimer's Association","keywords":"Artificial intelligence; Computer science; Frontotemporal dementia; Feature (linguistics); Pattern recognition (psychology); Machine learning; Perceptron; Frontotemporal lobar degeneration; Neuroimaging; Extractor; Dementia; Artificial neural network; Disease; Psychology; Medicine; Neuroscience; Pathology","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":[0.0002351658,0.0001434374,0.0002614372,0.0003486559,0.0001308043,0.00004931335,0.00009437201,0.00006223356,0.000008866719],"category_scores_gemma":[0.0004260006,0.0001330052,0.00007924331,0.0004060846,0.0000582507,0.0002031248,0.00003867915,0.0003010154,0.000005945745],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008357314,"about_ca_system_score_gemma":0.0001777763,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001683593,"about_ca_topic_score_gemma":0.000001385584,"domain_scores_codex":[0.9988532,0.00006759475,0.0003771595,0.0001744029,0.0002126327,0.0003150229],"domain_scores_gemma":[0.9993916,0.0001189778,0.00007182171,0.000143668,0.0002063684,0.00006759958],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008555246,0.0005624476,0.8750731,0.0007059151,0.00019718,0.00001076663,0.001922883,0.0003414338,0.0006611467,0.00151697,0.01963089,0.09852168],"study_design_scores_gemma":[0.004643283,0.0005194385,0.2120872,0.00008516289,0.0001034226,0.000002393271,0.003323603,0.7575185,0.000828952,0.0002648203,0.02048933,0.0001338728],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4945655,0.0001832282,0.4788864,0.001657051,0.0006568665,0.002815044,0.00001082698,0.0001368661,0.02108822],"genre_scores_gemma":[0.9670588,0.0003592313,0.0277935,0.001326295,0.00004269971,0.000307153,0.0001380508,0.00002319747,0.002951047],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7571771,"threshold_uncertainty_score":0.5423794,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01216786907003842,"score_gpt":0.2899944390999336,"score_spread":0.2778265700298951,"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."}}