{"id":"W3002763863","doi":"10.3389/fnins.2020.00015","title":"Shape Information Improves the Cross-Cohort Performance of Deep Learning-Based Segmentation of the Hippocampus","year":2020,"lang":"en","type":"article","venue":"Frontiers in Neuroscience","topic":"Medical Image Segmentation Techniques","field":"Computer Science","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute on Aging; National Institute of Biomedical Imaging and Bioengineering; Canadian Institutes of Health Research; National Institutes of Health; IXICO; Servier; Karolinska Institutet; Vetenskapsrådet; Eisai; H. Lundbeck A/S; Genentech; Stiftelsen för Strategisk Forskning; Northern California Institute for Research and Education; Stockholms Läns Landsting; Pfizer; Biogen; BioClinica; F. Hoffmann-La Roche; University of Southern California; European Commission; Eli Lilly and Company; U.S. Department of Defense; Meso Scale Diagnostics; Alzheimer's Disease Neuroimaging Initiative; Novartis Pharmaceuticals Corporation; Bristol-Myers Squibb; Alzheimer's Association; Foundation for the National Institutes of Health","keywords":"Segmentation; Context (archaeology); Computer science; Artificial intelligence; Cohort; Deep learning; Neuroimaging; Artificial neural network; Pattern recognition (psychology); Machine learning; Medicine; Psychology; Neuroscience; Pathology","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":[0.000482716,0.00008644736,0.000115853,0.0000745378,0.0001234851,0.00008379744,0.001413398,0.00003086598,0.000003179543],"category_scores_gemma":[0.0005736584,0.00005617589,0.000042327,0.001041441,0.0005128204,0.001322393,0.0001944151,0.0002033,9.355928e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000286014,"about_ca_system_score_gemma":0.00008256669,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005660071,"about_ca_topic_score_gemma":2.803716e-7,"domain_scores_codex":[0.9985464,0.0001129466,0.000390594,0.0001972113,0.0005940602,0.0001588448],"domain_scores_gemma":[0.9991782,0.00004806738,0.0003579572,0.000281582,0.00008886868,0.00004529183],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003174603,0.00004497432,0.4532231,0.000153171,0.000002543405,7.084131e-7,0.002909925,0.01315774,0.07761026,0.00007341849,0.0004923798,0.4523],"study_design_scores_gemma":[0.0001470702,0.000121194,0.1571912,0.00001166728,0.000001904278,5.640468e-7,0.00004776682,0.7047848,0.1375351,0.00006605643,0.00004393384,0.00004887152],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.318036,0.00001754626,0.6806058,0.00047393,0.0004583169,0.0003125499,8.530891e-7,0.00004073758,0.00005426955],"genre_scores_gemma":[0.9703194,0.00002043852,0.02726289,0.002357037,0.000007413942,0.00002364513,8.916499e-7,0.000002967022,0.000005293486],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.691627,"threshold_uncertainty_score":0.2626469,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008249968021111533,"score_gpt":0.2417768720453298,"score_spread":0.2335269040242183,"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."}}