{"id":"W3046390173","doi":"10.1017/s1431927620018693","title":"Deep Learning Convolutional Neural Networks for Pharmaceutical Tablet Defect Detection","year":2020,"lang":"en","type":"article","venue":"Microscopy and Microanalysis","topic":"Pharmaceutical Quality and Counterfeiting","field":"Medicine","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Object Research Systems (Canada)","funders":"","keywords":"Convolutional neural network; Computer science; Artificial intelligence; Action (physics); Deep learning; Pattern recognition (psychology); 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":[0.0003176304,0.0001865942,0.0003877888,0.00006017145,0.0003298845,0.00006408357,0.00005993426,0.0001150662,0.0001449103],"category_scores_gemma":[0.0001716085,0.0001796189,0.0003058574,0.0002979774,0.0001260693,0.00009184048,0.00005099195,0.0004611126,0.000009823568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000448653,"about_ca_system_score_gemma":0.00001889213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001259257,"about_ca_topic_score_gemma":0.000005961519,"domain_scores_codex":[0.9986554,0.0001035962,0.0003190839,0.0004096097,0.0001217342,0.0003905886],"domain_scores_gemma":[0.9992357,0.0001945109,0.00007739817,0.0000755634,0.0001026513,0.0003141184],"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":[0.002044091,0.00009589698,0.008107561,0.0003363515,0.0007987061,0.00001545028,0.0003324112,0.002162324,0.9428412,0.00006426407,0.0001703304,0.04303137],"study_design_scores_gemma":[0.002176955,0.0002546957,0.000423847,0.00002323591,0.001758596,0.00006516521,0.0002267907,0.8969114,0.07984011,0.00001676957,0.01807789,0.0002245942],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4867137,0.005673263,0.5047168,0.002258239,0.00008520334,0.0003351305,0.00001293463,0.0001085393,0.00009615403],"genre_scores_gemma":[0.9923773,0.0002066025,0.0008732448,0.005978059,0.0003596907,0.00002455425,0.00007586558,0.00002171288,0.00008291726],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.894749,"threshold_uncertainty_score":0.7324648,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04212779281063393,"score_gpt":0.3510686694337989,"score_spread":0.3089408766231649,"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."}}