{"id":"W4415150354","doi":"10.3390/jimaging11100360","title":"Lung Nodule Malignancy Classification Integrating Deep and Radiomic Features in a Three-Way Attention-Based Fusion Module","year":2025,"lang":"en","type":"article","venue":"Journal of Imaging","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; Health Sciences Centre; University of Toronto; Sunnybrook Health Science Centre; Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada; University of Toronto","keywords":"Pattern recognition (psychology); Deep learning; Convolutional neural network; Encoder; Feature extraction; Malignancy; Adenocarcinoma; Atypical adenomatous hyperplasia","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":[0.000848058,0.0001287064,0.0003266032,0.0005615731,0.0001125845,0.00005406619,0.0000970984,0.00004582329,0.00001499084],"category_scores_gemma":[0.0006010373,0.0001024038,0.0001022899,0.0002918733,0.00008174805,0.0001590333,0.00002739941,0.000695857,5.713563e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001676298,"about_ca_system_score_gemma":0.0001397224,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005605057,"about_ca_topic_score_gemma":0.00001332215,"domain_scores_codex":[0.998822,0.00006586216,0.0005154814,0.0001722587,0.0002361596,0.0001882122],"domain_scores_gemma":[0.9992055,0.0001309678,0.000290357,0.0001328729,0.0001488033,0.00009155144],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001105606,0.00009041013,0.8351858,0.0001909431,0.00003570901,0.0001735458,0.0001823153,0.0002226519,0.07349033,0.0004499335,0.0005584958,0.08930935],"study_design_scores_gemma":[0.001855709,0.00002490119,0.5461323,0.001578798,0.00007187366,0.000259243,0.0001980325,0.4490355,0.0003371018,0.0003422758,0.00009440548,0.00006984799],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8525332,0.003916678,0.1301556,0.01204599,0.000331861,0.0001491155,3.946769e-7,0.00001710964,0.0008499777],"genre_scores_gemma":[0.983785,0.00008233066,0.01510261,0.0007757403,0.0001302251,0.00000289781,0.000004099755,0.00001512243,0.0001020031],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4488128,"threshold_uncertainty_score":0.4175908,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005775253117694657,"score_gpt":0.2821782505826647,"score_spread":0.27640299746497,"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."}}