{"id":"W4306955959","doi":"10.1016/j.artmed.2022.102425","title":"Detecting 3D syndromic faces as outliers using unsupervised normalizing flow models","year":2022,"lang":"en","type":"article","venue":"Artificial Intelligence in Medicine","topic":"Face recognition and analysis","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"Alberta Children's Hospital; University of Calgary","funders":"National Institute of Dental and Craniofacial Research; National Institutes of Health; Canada Research Chairs; Calgary Foundation","keywords":"Interpretability; Computer science; Outlier; Artificial intelligence; Anomaly detection; Pattern recognition (psychology); Face (sociological concept); Unsupervised learning; Set (abstract data type); Machine learning; Data set","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":[0.00129372,0.0001924739,0.0003473221,0.0005950487,0.0005041777,0.0000779571,0.0008024588,0.00004581602,0.0005964455],"category_scores_gemma":[0.0002101766,0.0001889499,0.00008127922,0.001770473,0.0001125355,0.0005133817,0.000319549,0.0004234081,0.00005943203],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001911714,"about_ca_system_score_gemma":0.00009557427,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001838568,"about_ca_topic_score_gemma":0.0001912033,"domain_scores_codex":[0.9975168,0.0002304363,0.0006709032,0.0005204462,0.0006119389,0.0004495504],"domain_scores_gemma":[0.9990991,0.0001940529,0.000138812,0.000360649,0.0000801038,0.0001272949],"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":[0.00001466407,0.00006700122,0.0001085388,0.00001338324,0.00002428516,0.0001129943,0.007982441,0.7018189,0.007038465,0.003417771,0.00001305384,0.2793885],"study_design_scores_gemma":[0.00005512384,0.0001112149,0.000003539759,0.00006411529,0.00001823802,0.00005927441,0.01000771,0.9638854,0.004608454,0.02092079,0.00005878926,0.0002074011],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2731652,0.0002644401,0.7239075,0.00118,0.0005568803,0.0001642882,0.000001826173,0.0001132595,0.0006466098],"genre_scores_gemma":[0.9804764,0.00004612458,0.01858158,0.0007529252,0.00007282948,0.00002383822,0.000004434039,0.00001424449,0.00002760133],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7073113,"threshold_uncertainty_score":0.7705153,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.108275766305151,"score_gpt":0.326461303555088,"score_spread":0.218185537249937,"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."}}