{"id":"W4212905789","doi":"10.1101/2022.02.18.481050","title":"Deep learning accurately quantifies plasma cell percentages on CD138-stained bone marrow samples","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Digital Imaging for Blood Diseases","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"The Scarborough Hospital; University Health Network","funders":"","keywords":"Convolutional neural network; Artificial intelligence; Deep learning; Computer science; Segmentation; Software; Pattern recognition (psychology); Pathology; 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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0006799907,0.001071656,0.00089709,0.0007408338,0.0006861263,0.002653103,0.003044102,0.0002947787,0.0001619845],"category_scores_gemma":[0.0006271884,0.001216674,0.000428391,0.0009461218,0.0002082659,0.001036182,0.003841904,0.001551855,0.0002397317],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005263788,"about_ca_system_score_gemma":0.0007732756,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007528076,"about_ca_topic_score_gemma":0.000002786595,"domain_scores_codex":[0.9940472,0.000410805,0.0008217812,0.002310597,0.001217851,0.001191725],"domain_scores_gemma":[0.9953009,0.0003819921,0.000766259,0.002614997,0.0003892733,0.0005465088],"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":[0.0006812265,0.007369526,0.06642158,0.006382532,0.00164992,0.005634729,0.0008570706,0.06987248,0.7807945,0.04796153,0.01195646,0.0004184721],"study_design_scores_gemma":[0.00448877,0.00122422,0.1297219,0.00154552,0.0006690687,5.589305e-7,0.0003117919,0.08435193,0.6329829,0.0001203872,0.1341125,0.01047042],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9816293,0.002196012,0.007931467,0.0006807497,0.002798493,0.001049589,0.0003178568,0.003103968,0.0002925733],"genre_scores_gemma":[0.9848449,0.0002165585,0.01374364,0.0003662705,0.0002251748,0.0002923817,0.000003560697,0.0002091084,0.00009842138],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1478115,"threshold_uncertainty_score":0.9990283,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02551472015769561,"score_gpt":0.2400709050783664,"score_spread":0.2145561849206708,"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."}}