{"id":"W4409205157","doi":"10.1186/s40658-025-00750-7","title":"Impact of harmonization and oversampling methods on radiomics analysis of multi-center imbalanced datasets: application to PET-based prediction of lung cancer subtypes","year":2025,"lang":"en","type":"article","venue":"EJNMMI Physics","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Science and Technology Planning Project of Guangdong Province; Natural Science Foundation of Inner Mongolia; Natural Sciences and Engineering Research Council of Canada; National Natural Science Foundation of China","keywords":"Radiomics; Oversampling; Lung cancer; Harmonization; Center (category theory); Medicine; Medical physics; Computer science; Oncology; Radiology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01429,0.0008378752,0.0008889947,0.001010173,0.0005396758,0.000814445,0.0006507111,0.0004914948,0.0003349673],"category_scores_gemma":[0.01217328,0.000226859,0.0009905918,0.0007274224,0.0006330554,0.0007080063,0.001118231,0.0005921114,0.0001239141],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004681868,"about_ca_system_score_gemma":0.000767175,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00177096,"about_ca_topic_score_gemma":0.001865737,"domain_scores_codex":[0.9963183,0.002152621,0.0002585226,0.0007198326,0.0003888627,0.000161758],"domain_scores_gemma":[0.9915978,0.004824556,0.001077552,0.001480237,0.0008452744,0.0001745131],"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":[0.00255612,0.0009187901,0.2551051,0.0002342761,0.001580182,0.0001832632,0.0005666623,0.3180896,0.02385779,0.0009031435,0.00181096,0.3941941],"study_design_scores_gemma":[0.0001154593,0.001300162,0.07756352,0.00004169709,0.0004993718,0.0001991749,0.0002583951,0.8930724,0.02397048,0.001447411,0.001469015,0.0000628498],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8699022,0.0008237949,0.1275674,0.0001925818,0.00005631151,0.0001592172,0.0002168855,0.0005847215,0.0004969689],"genre_scores_gemma":[0.9540536,0.0001160254,0.0449718,0.00006177786,0.00003566105,0.00007449517,0.0004952069,0.00005144102,0.000139892],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01429,"threshold_uncertainty_score":0.07557368,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01484881667563729,"score_gpt":0.3957764222543597,"score_spread":0.3809276055787224,"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."}}