{"id":"W4409105075","doi":"10.1093/bioadv/vbaf021","title":"Transfer learning improves performance in volumetric electron microscopy organelle segmentation across tissues","year":2024,"lang":"en","type":"article","venue":"Bioinformatics Advances","topic":"Advanced Electron Microscopy Techniques and Applications","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Princess Margaret Cancer Centre; Lunenfeld-Tanenbaum Research Institute; Vector Institute; University of Toronto; University Health Network","funders":"","keywords":"Computer science; Transfer of learning; Segmentation; Benchmark (surveying); Artificial intelligence; Annotation; Pattern recognition (psychology); Deep learning; Identification (biology); Organelle; Endoplasmic reticulum; Image segmentation; Volume (thermodynamics); Microscopy; Biology; Pathology; Physics","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":[0.0001447047,0.0001798204,0.0001321845,0.00007074428,0.0001390941,0.0000868989,0.0001568465,0.00009852593,0.000007852701],"category_scores_gemma":[0.00001320717,0.0001717292,0.00004734861,0.0004575388,0.00006072476,0.00008613707,0.00003982496,0.000192717,0.00002511642],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005173354,"about_ca_system_score_gemma":0.00004158217,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005051766,"about_ca_topic_score_gemma":0.0000189691,"domain_scores_codex":[0.9989334,0.00001202735,0.0003264866,0.0002382055,0.0001026467,0.0003872047],"domain_scores_gemma":[0.9997125,0.00001219535,0.00004088621,0.0001572106,0.00003669825,0.00004052498],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0000228253,0.00002094844,0.0005130465,0.0001716149,0.000009802821,4.118997e-7,0.000257936,0.0006735248,0.9063895,0.00006365598,0.00007137151,0.09180537],"study_design_scores_gemma":[0.0001490271,0.0003286827,0.000120853,0.00003865013,0.000005340586,0.000008807961,0.0002519879,0.003342368,0.8273247,0.00003493279,0.1681974,0.0001972828],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9276978,0.009976396,0.06143824,0.00004453401,0.00008045277,0.0003407348,0.00001628623,0.0001029571,0.0003025802],"genre_scores_gemma":[0.962544,0.01486663,0.02115281,0.00008411119,0.00008408386,0.0001120906,0.0002565984,0.00003627486,0.000863424],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.168126,"threshold_uncertainty_score":0.7002913,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.004681615284073407,"score_gpt":0.3218662588925711,"score_spread":0.3171846436084977,"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."}}