{"id":"W4416913708","doi":"10.3390/info16121058","title":"Beyond Accuracy: Explainable Deep Learning for Alzheimer’s Disease Detection Using Structural MRI Data","year":2025,"lang":"en","type":"article","venue":"Information","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Wilfrid Laurier University","funders":"Canadian Institutes of Health Research; National Institutes of Health; H. Lundbeck A/S; Northern California Institute for Research and Education; Servier; BioClinica; University of Southern California; Bristol-Myers Squibb; Eli Lilly and Company; Biogen; Alliance de recherche numérique du Canada; Eisai; Alzheimer's Association; U.S. Department of Defense","keywords":"Interpretability; Deep learning; Convolutional neural network; Neuroimaging; Magnetic resonance imaging; Kernel (algebra); Artificial neural network; Limit (mathematics)","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0004327215,0.0001351166,0.0001117269,0.0002557623,0.0006734896,0.0006204806,0.0009527944,0.00005925381,0.000007838326],"category_scores_gemma":[0.0006585294,0.0001404248,0.00004166029,0.0005305641,0.00002873696,0.01209516,0.0004616863,0.0001342346,0.00003787872],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001118451,"about_ca_system_score_gemma":0.0001386316,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001258629,"about_ca_topic_score_gemma":0.00003841018,"domain_scores_codex":[0.9987833,0.00004373101,0.0004055848,0.0002385434,0.0002085022,0.0003202785],"domain_scores_gemma":[0.9985652,0.0001827624,0.0002111104,0.0007118928,0.0002500819,0.00007896512],"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":[0.00009382689,0.00001395593,0.000424209,0.0001155379,0.00004969706,0.000002024542,0.002265344,0.2117326,0.0004031997,0.0561144,0.0003182035,0.728467],"study_design_scores_gemma":[0.0001194223,0.00002292905,0.0003925087,0.00001972276,0.00002813915,0.000002305778,0.0004645466,0.9676308,0.007118659,0.01232792,0.01172344,0.0001496073],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01650762,0.0001519596,0.9809203,0.0003368635,0.000724244,0.0005127473,0.000006908537,0.0001949391,0.0006444252],"genre_scores_gemma":[0.9700718,0.00001088601,0.02928831,0.0003448359,0.0000665947,0.0000421704,0.0001373586,0.000005835153,0.0000321901],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9535642,"threshold_uncertainty_score":0.8768697,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04876642285899911,"score_gpt":0.3286879308589641,"score_spread":0.279921507999965,"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."}}