{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002538936,0.0000991639,0.0004007482,0.0002424895,0.00002528409,0.000005102138,0.00006105386,0.00003159212,0.000004593782],"category_scores_gemma":[0.0001721595,0.00008856096,0.0001147684,0.0008407465,0.00006223477,0.00004064412,0.00002203602,0.0001141326,6.809852e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009790055,"about_ca_system_score_gemma":0.00008367909,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002744997,"about_ca_topic_score_gemma":0.000002583916,"domain_scores_codex":[0.9992147,0.00005133252,0.0002961032,0.0002020391,0.0001411407,0.00009469694],"domain_scores_gemma":[0.999211,0.0001333332,0.0002203477,0.0002607945,0.0001211505,0.00005337999],"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":[0.0003655854,0.0005227459,0.6555638,0.0003366407,0.001567213,3.547817e-7,0.0002532969,0.1460033,0.1579997,0.0002544419,0.0002956223,0.0368373],"study_design_scores_gemma":[0.001255564,0.00006670916,0.2215386,0.0001615086,0.0009283177,1.386209e-7,0.000010032,0.7637384,0.01220772,0.00001166193,0.00003914721,0.00004220979],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4121158,0.00006891155,0.5871289,0.00008594168,0.00003680304,0.000185789,0.000355514,0.000009193753,0.00001310513],"genre_scores_gemma":[0.9694698,0.00006806316,0.02938129,0.0001145945,0.00003033128,0.00001190236,0.0009076297,0.00001071906,0.000005614011],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6177351,"threshold_uncertainty_score":0.3611411,"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."}}