{"id":"W3113100858","doi":"10.1016/j.heliyon.2020.e05652","title":"Improving Alzheimer's stage categorization with Convolutional Neural Network using transfer learning and different magnetic resonance imaging modalities","year":2020,"lang":"en","type":"article","venue":"Heliyon","topic":"Brain Tumor Detection and Classification","field":"Neuroscience","cited_by":101,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute on Aging; Eisai Incorporated; National Institute of Biomedical Imaging and Bioengineering; Canadian Institutes of Health Research; Genentech; National Institutes of Health; Takeda Pharmaceutical Company; IXICO; H. Lundbeck A/S; Servier; U.S. Department of Defense; Eli Lilly and Company; Université de Bordeaux; Centre National de la Recherche Scientifique; Northern California Institute for Research and Education; Alzheimer's Association; Fujirebio US; Pfizer; BioClinica; Biogen; Novartis Pharmaceuticals Corporation; Roche; Merck; Alzheimer's Drug Discovery Foundation; GE Healthcare; Alzheimer's Disease Neuroimaging Initiative; Meso Scale Diagnostics; AbbVie; Foundation for the National Institutes of Health","keywords":"Magnetic resonance imaging; Categorization; Convolutional neural network; Transfer of learning; Functional magnetic resonance imaging; Modalities; Stage (stratigraphy); Psychology; Artificial intelligence; Computer science; Neuroscience; Medicine; Radiology; Biology; Sociology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001274358,0.001327948,0.0007477846,0.001738791,0.0004093426,0.0007511922,0.001084929,0.001215108,0.001516582],"category_scores_gemma":[0.001998532,0.0002854872,0.00110562,0.0007927533,0.00033237,0.00126635,0.0008789632,0.001008003,0.0006149784],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001022254,"about_ca_system_score_gemma":0.0009116213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0102078,"about_ca_topic_score_gemma":0.007195815,"domain_scores_codex":[0.9995645,0.00006985281,0.00003145814,0.0001446734,0.00008806473,0.0001014937],"domain_scores_gemma":[0.9991787,0.0002896291,0.00008894195,0.0001221343,0.000259342,0.00006123819],"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":[0.000648229,0.0007904068,0.01357503,0.0001582712,0.0004470671,0.0002301379,0.0001231753,0.2481394,0.01945416,0.001075845,0.006854983,0.7085034],"study_design_scores_gemma":[0.000009765427,0.00006889848,0.001598032,0.000009707727,0.00004591655,0.00004010374,0.00001504255,0.9914214,0.005486721,0.0009485309,0.000343755,0.00001212167],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5590176,0.002607185,0.4246791,0.0006346031,0.0003242639,0.0002490808,0.000779254,0.006404276,0.005304665],"genre_scores_gemma":[0.9302187,0.0002697525,0.06521992,0.0001676489,0.00007157894,0.00006418896,0.001227644,0.00008686944,0.002673652],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0102078,"threshold_uncertainty_score":0.02029675,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03743440093654918,"score_gpt":0.23904008005585,"score_spread":0.2016056791193009,"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."}}