{"id":"W4413801790","doi":"10.1016/j.artmed.2025.103245","title":"Privacy-preserving federated transfer learning for enhanced liver lesion segmentation in PET–CT imaging","year":2025,"lang":"en","type":"article","venue":"Artificial Intelligence in Medicine","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Dalhousie University","funders":"Natural Science Foundation of Zhejiang Province; Science and Technology Program of Suzhou","keywords":"Computer science; Transfer of learning; Segmentation; Artificial intelligence; Computer vision; Medical imaging; Pattern recognition (psychology); Medical physics; Medicine","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001569244,0.00022928,0.0005081291,0.0006506735,0.0001654399,0.00004005669,0.0001921033,0.00003564808,0.0001921843],"category_scores_gemma":[0.003178912,0.0002038212,0.00007301967,0.0008583592,0.0001793715,0.0001645953,0.00005502488,0.0008410561,0.00001008998],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002128208,"about_ca_system_score_gemma":0.0001384575,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001123927,"about_ca_topic_score_gemma":0.0002546392,"domain_scores_codex":[0.9976491,0.0001516811,0.0008906617,0.0005059267,0.0003382486,0.0004643564],"domain_scores_gemma":[0.9988647,0.0006095371,0.00006992578,0.0001964042,0.0001539604,0.0001054455],"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":[0.0006966603,0.0002591517,0.0159273,0.0005165727,0.00003418832,0.0004581073,0.003070187,0.006194953,0.2812482,0.002540402,0.0002943719,0.6887599],"study_design_scores_gemma":[0.0012036,0.000263947,0.004651371,0.003576803,0.0000847158,0.00005658708,0.005975305,0.8590282,0.117584,0.006106677,0.001188089,0.0002806778],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6021866,0.0004971253,0.3836471,0.01015804,0.0004095356,0.0008825988,4.032206e-7,0.00007354323,0.002145036],"genre_scores_gemma":[0.9949505,0.0004661503,0.00259366,0.001308227,0.0001685448,0.00009129055,0.00004973313,0.00002553611,0.0003463526],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8528333,"threshold_uncertainty_score":0.8311587,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03924530443730946,"score_gpt":0.3719088386046395,"score_spread":0.33266353416733,"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."}}