{"id":"W4413900793","doi":"10.1101/2025.08.27.672636","title":"SPAGHETTI leverages massive H&E morphological models for phase contrast microscopy images with a generative deep learning approach","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Digital Holography and Microscopy","field":"Physics and Astronomy","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; University of Toronto","funders":"Social Sciences and Humanities Research Council; Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Princess Margaret Cancer Foundation","keywords":"Phase contrast microscopy; Contrast (vision); Generative grammar; Microscopy; Artificial intelligence; Computer science; Morphology (biology); Optics; Biology; Physics; Zoology","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003434383,0.0009727335,0.001048259,0.0002726985,0.0004910408,0.0006089939,0.0006217338,0.0004115792,0.00004573241],"category_scores_gemma":[0.00002902276,0.0008802108,0.0004195572,0.0004011352,0.0003889238,0.0003434257,0.0003773964,0.001301443,0.000008849833],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007926066,"about_ca_system_score_gemma":0.0004210365,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003908722,"about_ca_topic_score_gemma":6.192977e-7,"domain_scores_codex":[0.9965045,0.0001961039,0.0005559183,0.001604371,0.0002336672,0.0009054715],"domain_scores_gemma":[0.9977311,0.0001546173,0.0005421573,0.0006884199,0.0006221985,0.0002615393],"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":[0.0009290748,0.002635739,0.003799413,0.0005942702,0.002085721,0.00005087911,0.0001180213,0.02378162,0.9523361,0.01208321,0.001514067,0.00007191078],"study_design_scores_gemma":[0.00922416,0.0007815221,0.001097044,0.0009169083,0.0009856561,9.075659e-8,0.00011509,0.007833552,0.9690055,0.0002771409,0.006252632,0.003510711],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2983075,0.0007570152,0.6950996,0.00005369892,0.0002561085,0.001730375,0.003189846,0.0002525867,0.0003532528],"genre_scores_gemma":[0.9023286,0.00003700484,0.09558383,0.0001216732,0.0003613548,0.001336123,0.00002405696,0.0001015537,0.0001057678],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6040211,"threshold_uncertainty_score":0.9993649,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01799616644049577,"score_gpt":0.2564815593718673,"score_spread":0.2384853929313715,"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."}}