{"id":"W4280506572","doi":"10.1177/09544097221093486","title":"Use of deep convolutional neural networks and change detection technology for railway track inspections","year":2022,"lang":"en","type":"article","venue":"Proceedings of the Institution of Mechanical Engineers Part F Journal of Rail and Rapid Transit","topic":"Infrastructure Maintenance and Monitoring","field":"Engineering","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Pavemetrics Systems (Canada)","funders":"Federal Railroad Administration","keywords":"Change detection; Computer science; Fastener; Convolutional neural network; Field (mathematics); Software deployment; Process (computing); Track (disk drive); Artificial intelligence; Artificial neural network; Machine vision; Machine learning; Engineering; Software engineering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005468873,0.0004252061,0.0001912022,0.0006940805,0.0001971329,0.0005489073,0.0005284305,0.0005367366,0.0008190707],"category_scores_gemma":[0.001082412,0.0002357556,0.0002575881,0.0005939518,0.0002848868,0.0006348251,0.0003782106,0.0004991234,0.0002223237],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001343533,"about_ca_system_score_gemma":0.0006200139,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0129405,"about_ca_topic_score_gemma":0.01708115,"domain_scores_codex":[0.9996527,0.00003906193,0.00001736405,0.00007891848,0.0001771775,0.00003468737],"domain_scores_gemma":[0.9994954,0.0001497477,0.00007995214,0.00006748619,0.0001873658,0.00002000716],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002496967,0.0002681242,0.01158286,0.0001169171,0.00007909433,0.0001958942,0.0001229121,0.2174331,0.1022004,0.002986325,0.003196763,0.661568],"study_design_scores_gemma":[0.000005784146,0.00009252886,0.005211044,0.00001416391,0.00002015874,0.00006941223,0.00001615346,0.9509424,0.04081588,0.0008050229,0.001991658,0.0000157185],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2836058,0.0006910095,0.7021461,0.0004711673,0.0001028757,0.00007751343,0.0003708098,0.005385004,0.007149653],"genre_scores_gemma":[0.8343732,0.0002716935,0.1616719,0.0001429059,0.00001602062,0.0000344129,0.0003806163,0.00005704099,0.003052163],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0129405,"threshold_uncertainty_score":0.02573037,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01464035208344411,"score_gpt":0.1936580649380174,"score_spread":0.1790177128545733,"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."}}