{"id":"W4409853754","doi":"10.1186/s12938-025-01376-7","title":"The use of a convolutional neural network to automate radiologic scoring of computed tomography of paranasal sinuses","year":2025,"lang":"en","type":"article","venue":"BioMedical Engineering OnLine","topic":"Sinusitis and nasal conditions","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; Toronto Public Health; St. Michael's Hospital; Queen's University; North York General Hospital","funders":"","keywords":"Paranasal sinuses; Medicine; Segmentation; Frontal sinus; Convolutional neural network; Sinus (botany); Ethmoid sinus; Sørensen–Dice coefficient; Radiology; Paranasal Sinus Diseases; Coronal plane; Artificial intelligence; Thresholding; Nuclear medicine; Computer science; Image segmentation; Surgery","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":[0.001059544,0.0007210908,0.0002665849,0.0008353752,0.0002498154,0.000504192,0.0007896503,0.000554625,0.0009619133],"category_scores_gemma":[0.002202634,0.0003211808,0.0004156441,0.0003725203,0.0002327439,0.0004775053,0.0004890118,0.0004323363,0.000405512],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001043884,"about_ca_system_score_gemma":0.001194337,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01160103,"about_ca_topic_score_gemma":0.01310424,"domain_scores_codex":[0.9995795,0.00006030767,0.00002836761,0.0001064008,0.0001630938,0.00006230119],"domain_scores_gemma":[0.9992002,0.0002035543,0.0001244161,0.00007930084,0.0003608243,0.00003168637],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004939383,0.0003348464,0.02099344,0.0001572393,0.0002621373,0.0003777936,0.0000719561,0.2686933,0.1236905,0.001458435,0.003816352,0.57965],"study_design_scores_gemma":[0.00000821008,0.00006615157,0.00342622,0.000008335378,0.00002845439,0.00009977213,0.000005748946,0.9736728,0.02178969,0.0002512283,0.0006334748,0.000009905438],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2827937,0.0004368556,0.7074599,0.0003324081,0.0001167348,0.0002305765,0.0004372286,0.005262502,0.002930083],"genre_scores_gemma":[0.7601965,0.0001855232,0.2368373,0.0001305809,0.00003047606,0.000115399,0.0006580611,0.00007735224,0.001768816],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01160103,"threshold_uncertainty_score":0.023067,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02978224768336767,"score_gpt":0.2777141777661076,"score_spread":0.2479319300827399,"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."}}