{"id":"W4416402424","doi":"10.3389/fdgth.2025.1659366","title":"CharMark: character-level Markov modeling for interpretable linguistic biomarkers of cognitive decline","year":2025,"lang":"en","type":"article","venue":"Frontiers in Digital Health","topic":"Neurobiology of Language and Bilingualism","field":"Neuroscience","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Pittsburgh; National Institute on Aging; National Science Foundation","keywords":"Cognitive decline; Discriminative model; Dementia; Cognition; Interpretation (philosophy); Hidden Markov model; Character (mathematics); Logistic regression; Computational linguistics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00167237,0.0009539165,0.0008223143,0.0008461331,0.0005394758,0.001371455,0.001589796,0.001226631,0.003494945],"category_scores_gemma":[0.007644835,0.0005730576,0.00147808,0.000586613,0.0006830338,0.001002414,0.001437972,0.002285755,0.0008909493],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001242499,"about_ca_system_score_gemma":0.00150442,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01389066,"about_ca_topic_score_gemma":0.01583263,"domain_scores_codex":[0.9993748,0.0002906661,0.0000303201,0.0001605667,0.00008849282,0.00005509605],"domain_scores_gemma":[0.9966772,0.002681309,0.0002280122,0.0001374095,0.0001827587,0.00009344466],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002201386,0.000111593,0.004293131,0.0001318938,0.0001500359,0.0001471259,0.0001991751,0.8910468,0.001624499,0.03624158,0.003963107,0.06187078],"study_design_scores_gemma":[0.000003761211,0.00000873082,0.0001221237,0.000005992421,0.000005359834,0.000007776698,0.000005057514,0.9886664,0.00009273789,0.01072023,0.0003577814,0.000004071333],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01635161,0.0003901116,0.9785567,0.0006245861,0.00007623782,0.00007913837,0.001086555,0.001729317,0.001105817],"genre_scores_gemma":[0.6169276,0.000907049,0.3680758,0.000713,0.0002984619,0.001057198,0.004921662,0.0004978611,0.006601334],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01389066,"threshold_uncertainty_score":0.0276196,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03546149579726589,"score_gpt":0.3344673724246631,"score_spread":0.2990058766273972,"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."}}