{"id":"W3119061771","doi":"10.1007/s12021-020-09499-z","title":"Generating Longitudinal Atrophy Evaluation Datasets on Brain Magnetic Resonance Images Using Convolutional Neural Networks and Segmentation Priors","year":2021,"lang":"en","type":"article","venue":"Neuroinformatics","topic":"Cell Image Analysis Techniques","field":"Biochemistry, Genetics and Molecular Biology","cited_by":17,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"National Center for Research Resources; Canadian Institutes of Health Research; Fundació la Marató de TV3; Agència de Gestió d'Ajuts Universitaris i de Recerca; Ministerio de Ciencia, Innovación y Universidades; National Institute on Aging","keywords":"Sørensen–Dice coefficient; Segmentation; Artificial intelligence; Magnetic resonance imaging; Atrophy; Convolutional neural network; Pattern recognition (psychology); Similarity (geometry); Computer science; Ground truth; Neuroimaging; Brain size; Image segmentation; Neuroscience; Medicine; Pathology; Psychology; Radiology; Image (mathematics)","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":[0.0002480846,0.0001440169,0.000114541,0.00004566625,0.0001363431,0.0001074072,0.00007762519,0.00006427345,0.00001923709],"category_scores_gemma":[0.0002813616,0.0001541958,0.00004683042,0.0001206201,0.00006369904,0.00003100506,0.0001252033,0.0001100566,0.00000150676],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000256927,"about_ca_system_score_gemma":0.00005974271,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005206505,"about_ca_topic_score_gemma":0.000009713254,"domain_scores_codex":[0.998902,0.0001139868,0.0003060411,0.0002373836,0.0002667781,0.0001737888],"domain_scores_gemma":[0.9993244,0.00003971462,0.0001315538,0.0003060414,0.0001494062,0.00004884878],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007363207,0.000086907,0.005589885,0.0000712456,0.00003944304,0.00003435087,0.00008715496,0.09852998,0.8124207,0.00003495729,0.02025229,0.06277947],"study_design_scores_gemma":[0.0004202477,0.0001780732,0.003826274,0.00001510034,0.00006605104,0.0001120019,0.00003880304,0.9365863,0.05755997,0.00001030474,0.001014865,0.0001720774],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9752235,0.001610686,0.02259255,0.0001102161,0.00004888868,0.0002304028,0.00003781414,0.00001803582,0.0001279503],"genre_scores_gemma":[0.963488,0.000350332,0.03040407,0.002250444,0.0002556512,0.00001450341,0.00312587,0.000029062,0.00008211009],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8380563,"threshold_uncertainty_score":0.6287924,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01782677807473911,"score_gpt":0.2949912237290832,"score_spread":0.2771644456543441,"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."}}