{"id":"W2765527079","doi":"10.1136/gutjnl-2017-314547","title":"Real-time differentiation of adenomatous and hyperplastic diminutive colorectal polyps during analysis of unaltered videos of standard colonoscopy using a deep learning model","year":2017,"lang":"en","type":"article","venue":"Gut","topic":"Colorectal Cancer Screening and Detection","field":"Medicine","cited_by":627,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal; Vancouver General Hospital; University of British Columbia","funders":"Boston Scientific Corporation","keywords":"Diminutive; Hyperplastic Polyp; Colonoscopy; Medicine; Narrow-band imaging; Artificial intelligence; Colorectal Polyp; Predictive value; Radiology; Endoscopy; Internal medicine; Gastroenterology; Colorectal cancer; Computer science; Cancer","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":[0.0001074545,0.000104681,0.0005615588,0.0002970164,0.0001411572,0.00001181056,0.00005289751,0.00007124739,0.0000136401],"category_scores_gemma":[0.0002266939,0.0001017404,0.0001108873,0.0001665262,0.0001202902,0.00007747892,0.00005954173,0.00009153789,1.760368e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000917678,"about_ca_system_score_gemma":0.00006288537,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006064929,"about_ca_topic_score_gemma":0.0001252045,"domain_scores_codex":[0.9991247,0.00003713786,0.000308887,0.0001837342,0.0002209144,0.0001246068],"domain_scores_gemma":[0.9989753,0.00007730257,0.000547338,0.0001739961,0.0001685529,0.0000575292],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.01403398,0.0000582442,0.1433469,0.0002182432,0.0008411157,0.000003909151,0.001733509,0.01196839,0.8269523,0.000006837304,3.28751e-7,0.0008362132],"study_design_scores_gemma":[0.0009093274,0.001345865,0.4165492,0.0001370204,0.001375842,0.000007129664,0.00009955924,0.4803461,0.09915435,0.00001502744,7.206425e-8,0.00006047258],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9952942,0.0000693824,0.004271131,0.00000504387,0.00002803536,0.0001561872,0.00003054393,0.00001843919,0.0001270497],"genre_scores_gemma":[0.9985863,0.0000337657,0.001306588,6.681238e-7,0.00001645089,0.000004539983,0.000009779765,0.0000117055,0.000030217],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.727798,"threshold_uncertainty_score":0.4148853,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01600855800629609,"score_gpt":0.283871320981928,"score_spread":0.2678627629756319,"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."}}