{"id":"W4412714137","doi":"10.1016/j.phro.2025.100817","title":"Image-based mandibular and maxillary parcellation and annotation using computed tomography (IMPACT): a deep learning-based clinical tool for orodental dose estimation and osteoradionecrosis assessment","year":2025,"lang":"en","type":"article","venue":"Physics and Imaging in Radiation Oncology","topic":"Dental Radiography and Imaging","field":"Dentistry","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Princess Margaret Cancer Centre","funders":"National Center for Advancing Translational Sciences; Center for Clinical and Translational Sciences, University of Texas Health Science Center at Houston; KWF Kankerbestrijding; Hitachi; National Cancer Institute; National Institutes of Health; National Aeronautics and Space Administration; National Institute of Dental and Craniofacial Research; University of Texas MD Anderson Cancer Center; Cancer Prevention and Research Institute of Texas; ZonMw; Baylor College of Medicine","keywords":"Computed tomography; Annotation; Osteoradionecrosis; Estimation; Artificial intelligence; Computer science; Medicine; Orthodontics; Dentistry; Radiation therapy; Radiology; Engineering","routes":{"ca_aff":true,"ca_fund":false,"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":[0.0006248966,0.0001985848,0.0003314613,0.0003648511,0.0002884697,0.0002674613,0.00004133384,0.0000879877,0.000002573821],"category_scores_gemma":[0.00006539248,0.0002199071,0.00008075839,0.0003628376,0.0002302718,0.0004656685,0.00003626621,0.0001994329,2.265662e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001247899,"about_ca_system_score_gemma":0.00007942313,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009925961,"about_ca_topic_score_gemma":0.00001408724,"domain_scores_codex":[0.9984992,0.0002749298,0.0004546484,0.0004352102,0.0001189388,0.0002170583],"domain_scores_gemma":[0.9989524,0.000526979,0.000262047,0.00009994456,0.00008691191,0.00007166197],"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":[0.000157671,0.0001318436,0.746415,0.000156066,0.00003618024,0.0000108177,0.00009724258,0.004938581,0.001891277,0.000199722,0.00002386631,0.2459418],"study_design_scores_gemma":[0.002505522,0.00009628153,0.4570088,0.00004299634,0.00006929709,0.00001436557,0.00004583154,0.5393986,0.00007788881,0.0005732332,0.0000540033,0.0001131886],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6796673,0.0009067174,0.3185974,0.0001551882,0.00016416,0.0004355206,0.00001726628,0.00003166726,0.00002481895],"genre_scores_gemma":[0.9655833,0.0001098794,0.03370151,0.0002556341,0.00004693479,0.0000330987,0.000250918,0.00001717822,0.000001570224],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5344599,"threshold_uncertainty_score":0.8967552,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01429160763060082,"score_gpt":0.3715181879923098,"score_spread":0.357226580361709,"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."}}