{"id":"W4413149186","doi":"10.1016/j.compmedimag.2025.102630","title":"Inference time correction based on confidence and uncertainty for improved deep-learning model performance and explainability in medical image classification","year":2025,"lang":"en","type":"article","venue":"Computerized Medical Imaging and Graphics","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"ASTER","funders":"Qatar National Research Fund; Qatar Foundation","keywords":"Inference; Artificial intelligence; Computer science; Deep learning; Machine learning; Image (mathematics); Confidence interval; Medical imaging; Pattern recognition (psychology); Computer vision; Statistics; Mathematics","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":[0.004755572,0.0008275187,0.001057168,0.0007430221,0.0005245851,0.001248658,0.002029512,0.00195663,0.002305551],"category_scores_gemma":[0.03102245,0.0007020546,0.0009383266,0.0006024739,0.0009930332,0.003207641,0.002349724,0.004647251,0.000278568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00120552,"about_ca_system_score_gemma":0.001735181,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005434448,"about_ca_topic_score_gemma":0.005149717,"domain_scores_codex":[0.9985117,0.0004506453,0.0001302627,0.0003940165,0.0003628071,0.0001506174],"domain_scores_gemma":[0.9787066,0.01637474,0.001057206,0.001774484,0.001675976,0.0004110542],"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":[0.001293277,0.0002245913,0.006184909,0.0001557201,0.0001599875,0.0002236861,0.0002017177,0.7341282,0.0106849,0.02943501,0.002391069,0.2149169],"study_design_scores_gemma":[0.000007038462,0.00001592141,0.0002242264,0.000005481837,0.000008284474,0.00001221136,0.000003460478,0.9933642,0.001647791,0.00462133,0.00008507793,0.000004981996],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07241815,0.0005825271,0.9244912,0.0008592897,0.000117507,0.00002347349,0.000135348,0.0008746535,0.0004978727],"genre_scores_gemma":[0.8658239,0.000211794,0.131559,0.0002307459,0.0001290837,0.00004345322,0.0003619063,0.000328835,0.001311297],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005434448,"threshold_uncertainty_score":0.02515018,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008764051783965858,"score_gpt":0.3013993181054609,"score_spread":0.2926352663214951,"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."}}