{"id":"W4413055840","doi":"10.1177/15578666251364292","title":"Enhanced Interpretable Neural Network Approach for Unified Batch Effect Mitigation and Disease Classification Using Cross-Cohort Microbiome Profiles","year":2025,"lang":"en","type":"article","venue":"Journal of Computational Biology","topic":"Oral microbiology and periodontitis research","field":"Dentistry","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University; University of Manitoba","funders":"","keywords":"Microbiome; Cohort; Artificial neural network; Disease; Artificial intelligence; Computer science; Machine learning; Computational biology; Biology; Mathematics; Medicine; Bioinformatics; Statistics; Internal medicine","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006376133,0.0001237807,0.0002995919,0.000163802,0.0002033682,0.00008308901,0.0001545302,0.0001511649,0.00002853147],"category_scores_gemma":[0.0002319046,0.0001044692,0.0001119914,0.0001844069,0.000260034,0.000153278,0.00005214087,0.0002045682,0.000002302648],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006491122,"about_ca_system_score_gemma":0.0001949721,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004464393,"about_ca_topic_score_gemma":9.562856e-7,"domain_scores_codex":[0.998743,0.0003074928,0.0004630322,0.0002172617,0.00005920108,0.0002100275],"domain_scores_gemma":[0.9986578,0.0004856355,0.0002810254,0.00007741193,0.0004224995,0.00007558567],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.002984941,0.0001200727,0.5574627,0.0003464624,0.0004325737,0.000006892234,0.00003611936,0.05733638,0.3704743,0.005094773,0.001907002,0.003797829],"study_design_scores_gemma":[0.002629424,0.0004244427,0.8169205,0.000135534,0.0001459564,0.0001593381,0.00001953051,0.1576737,0.005601109,0.01558639,0.0004645061,0.0002395527],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8089931,0.0005368054,0.189288,0.0001466352,0.0005447448,0.000351655,0.00006703909,0.000008246549,0.00006373425],"genre_scores_gemma":[0.9895566,0.000008621302,0.009501793,0.000138249,0.000195963,0.00001729912,0.0004072643,0.000007745013,0.0001664895],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3648732,"threshold_uncertainty_score":0.4260129,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01939396327145821,"score_gpt":0.3546994212512898,"score_spread":0.3353054579798316,"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."}}