{"id":"W4415271613","doi":"10.3390/math13203309","title":"Dataset-Aware Preprocessing for Hippocampal Segmentation: Insights from Ablation and Transfer Learning","year":2025,"lang":"en","type":"article","venue":"Mathematics","topic":"Medical Imaging and Analysis","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institutes of Health Research; Genentech; National Institutes of Health; Eisai; National Research Foundation of Korea; Chosun University; Ministry of Science and ICT, South Korea; BioClinica; U.S. Department of Defense; Alzheimer's Disease Neuroimaging Initiative; Northern California Institute for Research and Education; F. Hoffmann-La Roche; National Research Foundation; Bristol-Myers Squibb; Eli Lilly and Company; Biogen; Alzheimer's Association","keywords":"Segmentation; Preprocessor; Pipeline (software); Transfer of learning; Pattern recognition (psychology); Data pre-processing; Deep learning; HARP; Hausdorff distance; Perspective (graphical)","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":[0.00005364944,0.00006555882,0.0001093657,0.00004239569,0.00007084119,0.00005078268,0.00003637166,0.00003137347,0.00002124317],"category_scores_gemma":[0.00003633567,0.00005918129,0.00001643789,0.00008016962,0.00001432584,0.00007796238,0.000006423914,0.00006385604,0.000002865464],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001095849,"about_ca_system_score_gemma":0.000005996816,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002878672,"about_ca_topic_score_gemma":0.000002087501,"domain_scores_codex":[0.9996272,0.000006828415,0.0001449999,0.00009042385,0.00006640777,0.00006419098],"domain_scores_gemma":[0.99975,0.0001264349,0.000009691747,0.00007474666,0.0000139244,0.00002524646],"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":[0.00002939662,0.0004693544,0.007689874,0.02257103,0.002444556,0.00001746775,0.06452011,0.2030556,0.09013692,0.005615837,0.02673873,0.5767112],"study_design_scores_gemma":[0.0002706349,0.00000328889,0.00006265475,0.0001992546,0.0001228759,3.128862e-7,0.0007846257,0.9862446,0.004100132,0.006491281,0.001633939,0.0000863712],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.161643,0.0002822148,0.83754,0.00004934872,0.00003885581,0.00007128482,0.00003774621,0.0000784094,0.00025914],"genre_scores_gemma":[0.9792101,0.00002848893,0.01977192,0.00004957,0.00004108909,0.00002142089,0.0007297507,0.00001221389,0.0001354391],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8177681,"threshold_uncertainty_score":0.2413343,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0118388309922994,"score_gpt":0.2570011535026089,"score_spread":0.2451623225103095,"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."}}