{"id":"W4414561830","doi":"10.1093/dmfr/twaf069","title":"Single-step prediction of inferior alveolar nerve injury after mandibular third-molar extraction using contrastive learning and Bayesian auto-tuned deep learning model","year":2025,"lang":"en","type":"article","venue":"Dentomaxillofacial Radiology","topic":"Dental Radiography and Imaging","field":"Dentistry","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Ministry of Science and ICT, South Korea; National Research Foundation of Korea","keywords":"Deep learning; Bayesian probability; Focus (optics); Bayesian network; Chest radiograph; Convolutional neural network; Inferior alveolar nerve; Panoramic radiograph","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005070379,0.0003897223,0.0006791315,0.0008140766,0.0004085895,0.0001213179,0.0001566157,0.0004548711,0.00007010451],"category_scores_gemma":[0.0003450528,0.0004377282,0.0002799617,0.0005660469,0.0003850214,0.000670724,0.0001387253,0.0008244712,0.000007513001],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001941593,"about_ca_system_score_gemma":0.00005929861,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001725709,"about_ca_topic_score_gemma":0.00003336009,"domain_scores_codex":[0.9972736,0.0004099486,0.0007844839,0.0006646228,0.0002989956,0.0005683919],"domain_scores_gemma":[0.9989871,0.0001013091,0.0004023556,0.0002042238,0.0001664565,0.0001385279],"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":[0.000933928,0.0001219341,0.7338333,0.0001156955,0.0004056382,0.0002179483,0.0006583023,0.002284724,0.2438952,0.0003197173,0.00001459121,0.01719905],"study_design_scores_gemma":[0.002068784,0.0004691965,0.4600292,0.0002330911,0.000618965,0.0007586433,0.001018425,0.5223576,0.009258213,0.0006485397,0.001979566,0.0005597764],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7973531,0.001851036,0.1987505,0.00001336426,0.000904609,0.0003145861,0.00003676168,0.0001514086,0.0006246341],"genre_scores_gemma":[0.9974809,0.0001292705,0.001797242,0.00006245776,0.00009157707,0.00003604585,0.00006126973,0.00004360324,0.0002976184],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5200729,"threshold_uncertainty_score":0.9998075,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00883327976971053,"score_gpt":0.2716184993445473,"score_spread":0.2627852195748367,"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."}}