{"id":"W4293578078","doi":"10.1016/j.jpi.2022.100133","title":"A self-supervised contrastive learning approach for whole slide image representation in digital pathology","year":2022,"lang":"en","type":"article","venue":"Journal of Pathology Informatics","topic":"AI in cancer detection","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; McMaster University; University of Waterloo","funders":"Government of Ontario","keywords":"Computer science; Digital pathology; Artificial intelligence; Machine learning; Robustness (evolution); Supervised learning; Pattern recognition (psychology); Natural language processing; Artificial neural network","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":[0.001218159,0.0001152435,0.000303826,0.0003837026,0.0001776401,0.00009755965,0.0005223636,0.00007211009,0.000003395359],"category_scores_gemma":[0.0003310529,0.0001145097,0.0001113655,0.0003862153,0.0000502565,0.001702449,0.0002495747,0.0006553431,0.000002866798],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002551386,"about_ca_system_score_gemma":0.0001510255,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":5.572904e-7,"about_ca_topic_score_gemma":3.293277e-7,"domain_scores_codex":[0.9983202,0.0001532058,0.0008779872,0.0001126239,0.0002791376,0.0002569089],"domain_scores_gemma":[0.9984018,0.0002385103,0.0008872778,0.0001813655,0.0002400355,0.00005106194],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001222299,0.001673788,0.02496793,0.0006272295,0.0003457432,0.002717442,0.3039921,0.3826123,0.0107027,0.005671941,0.003626422,0.2618401],"study_design_scores_gemma":[0.003399989,0.001961118,0.001186601,0.00001252942,0.00002689768,0.008043659,0.01142156,0.9684747,0.0004374236,0.002925783,0.001878204,0.0002315649],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.165446,0.00003944999,0.833008,0.0001794418,0.0003838927,0.0002473547,0.000009510994,0.00003687391,0.0006495125],"genre_scores_gemma":[0.5819035,0.00001621491,0.4176402,0.0002448455,0.00007801501,0.00006496275,0.00001303373,0.00001079368,0.0000284309],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5858624,"threshold_uncertainty_score":0.4669569,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01397487569922593,"score_gpt":0.2580415608537546,"score_spread":0.2440666851545286,"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."}}