{"id":"W4414051561","doi":"10.1162/imag.a.158","title":"Deep-learning segmentation of the substantia nigra from multiparametric MRI: Application to Parkinson’s disease","year":2025,"lang":"en","type":"article","venue":"Imaging Neuroscience","topic":"Neurological disorders and treatments","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"London Health Sciences Centre; Western University","funders":"Avid Radiopharmaceuticals; Allergan; Genentech; H. Lundbeck A/S; Servier; Norges Forskningsråd; Sun Pharma; Voyager Therapeutics; Neurocrine Biosciences; Jazz Pharmaceuticals; Stiftelsen Kristian Gerhard Jebsen; Biogen; Celgene; Universitetet i Bergen; Verily Life Sciences; Teva Pharmaceutical Industries; Sanofi; Pfizer; Bristol-Myers Squibb; Eli Lilly and Company","keywords":"Substantia nigra; Pars compacta; Hyperintensity; Segmentation; Neuromelanin; Artificial neural network; Disease; Magnetic resonance imaging","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00005089198,0.00007929243,0.00009568845,0.0001070601,0.0001265748,0.00002493762,0.0001501004,0.000009829627,0.000003740893],"category_scores_gemma":[0.0003869596,0.00005463429,0.00004866175,0.001108006,0.00009172451,0.00005801646,0.00006611417,0.00009066251,0.000005327922],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002047995,"about_ca_system_score_gemma":0.00002739516,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007792701,"about_ca_topic_score_gemma":0.0000036079,"domain_scores_codex":[0.9991259,0.00004823694,0.0001360803,0.0003376554,0.0002153468,0.0001367719],"domain_scores_gemma":[0.9994955,0.00008168832,0.00006305789,0.0002437837,0.00003804586,0.00007790196],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00004906861,0.000129589,0.910426,0.000008051652,0.000001621381,0.000006183307,0.00004029577,0.0007474278,0.06977639,0.00002741626,0.00005988605,0.01872806],"study_design_scores_gemma":[0.0003744341,0.00002673323,0.9607617,0.00001963235,0.00004022288,3.516719e-7,0.00001898531,0.0326264,0.004721277,0.0003384989,0.001031675,0.00004013273],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9632419,0.0001855228,0.03227067,0.003373505,0.0001863673,0.0005281005,0.000005250289,0.00004156631,0.0001671302],"genre_scores_gemma":[0.9947796,0.00003230495,0.0004432894,0.004595155,0.000005642014,0.0000368205,0.000003270088,0.000004796425,0.00009913887],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06505512,"threshold_uncertainty_score":0.2227922,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01023717123824822,"score_gpt":0.2832032847057002,"score_spread":0.2729661134674519,"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."}}