{"id":"W4207027034","doi":"10.1101/2022.01.22.22269695","title":"Segmentation stability of human head and neck cancer medical images for radiotherapy applications under de-identification conditions: benchmarking data sharing and artificial intelligence use-cases","year":2022,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Medical Imaging and Analysis","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Institute of Biomedical Imaging and Bioengineering; National Institute of Dental and Craniofacial Research; University of Texas MD Anderson Cancer Center; University of Texas Health Science Center at Houston; Center for Clinical and Translational Sciences, University of Texas Health Science Center at Houston; Division of Mathematical Sciences; National Cancer Institute; National Institutes of Health; National Science Foundation","keywords":"Computer science; Artificial intelligence; Segmentation; Benchmarking; Medical imaging; Medical physics; Deep learning; Head and neck cancer; Medicine; Radiation therapy; Radiology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006831014,0.001313365,0.001090142,0.002165079,0.0006361821,0.00155089,0.002011574,0.001726578,0.001154502],"category_scores_gemma":[0.01415741,0.0004166537,0.001411227,0.001425341,0.001053279,0.001276092,0.001819436,0.0009276664,0.0007431422],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001614375,"about_ca_system_score_gemma":0.001265572,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01475323,"about_ca_topic_score_gemma":0.01185476,"domain_scores_codex":[0.9961521,0.0008613699,0.000448396,0.001299588,0.0009022027,0.0003363648],"domain_scores_gemma":[0.9944605,0.00212428,0.0004352639,0.00135254,0.001391144,0.0002362646],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003660095,0.001261655,0.05530271,0.001340589,0.001026219,0.0007339647,0.001006426,0.4960281,0.03588604,0.001149423,0.01540999,0.3871948],"study_design_scores_gemma":[0.000100323,0.0009481342,0.03260925,0.0001151266,0.0001739866,0.0005978436,0.0004664486,0.9145356,0.0450974,0.001160183,0.004136656,0.00005913994],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9574455,0.002175663,0.02880318,0.0004739651,0.0002371592,0.000325732,0.003024283,0.004923031,0.002591499],"genre_scores_gemma":[0.9498051,0.0004143398,0.03630969,0.0001335621,0.00004202897,0.000115297,0.0118731,0.0002517891,0.001055128],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01475323,"threshold_uncertainty_score":0.03612632,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1428809440247545,"score_gpt":0.4051903173518598,"score_spread":0.2623093733271054,"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."}}