{"id":"W3167649396","doi":"10.1007/s10334-021-00934-z","title":"Automatic knee cartilage and bone segmentation using multi-stage convolutional neural networks: data from the osteoarthritis initiative","year":2021,"lang":"en","type":"article","venue":"Magnetic Resonance Materials in Physics Biology and Medicine","topic":"Osteoarthritis Treatment and Mechanisms","field":"Medicine","cited_by":42,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo; McMaster University","funders":"Institute of Musculoskeletal Health and Arthritis; Natural Sciences and Engineering Research Council of Canada; Mitacs; Arthritis Society; Government of Ontario","keywords":"Segmentation; Cartilage; Convolutional neural network; Osteoarthritis; Artificial intelligence; Computer science; Knee cartilage; Pattern recognition (psychology); Sørensen–Dice coefficient; Image segmentation; Medicine; Anatomy; Articular cartilage; Pathology","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002383058,0.0001846428,0.0004541305,0.00002279508,0.0001162417,0.00001772884,0.00005863787,0.0001007981,0.0002720851],"category_scores_gemma":[0.0001160863,0.0001323281,0.00001373217,0.0001083855,0.0004493724,0.00009379586,0.0001632513,0.0001118176,0.000001701085],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002065869,"about_ca_system_score_gemma":0.00004841888,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001148883,"about_ca_topic_score_gemma":0.0001344814,"domain_scores_codex":[0.9986673,0.0002458844,0.0003480945,0.0004065171,0.00009784712,0.0002343766],"domain_scores_gemma":[0.9992296,0.0002393108,0.0001031353,0.0003163775,0.0000497092,0.00006193807],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.0003397627,0.0001518775,0.02793608,0.00009213157,0.00002449339,0.0004071913,0.001285139,0.000004039187,0.7045019,0.001009098,0.00006242256,0.2641858],"study_design_scores_gemma":[0.1249342,0.01536962,0.5547706,0.007615124,0.002085567,0.001629527,0.01159924,0.115222,0.1471982,0.01334998,0.004130423,0.002095483],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9526237,0.0452619,0.0002467069,0.0004995185,0.0004567226,0.0004696767,0.0003953311,0.00001930172,0.00002714213],"genre_scores_gemma":[0.9931031,0.001459647,0.001984684,0.000977147,0.0005558888,0.00004223496,0.001794823,0.00001569228,0.00006678195],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5573037,"threshold_uncertainty_score":0.5396183,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06602622815629056,"score_gpt":0.3261400846650832,"score_spread":0.2601138565087926,"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."}}