{"id":"W4385481139","doi":"10.1016/j.ibmed.2023.100103","title":"Toward enhanced free-living fall risk assessment: Data mining and deep learning for environment and terrain classification","year":2023,"lang":"en","type":"article","venue":"Intelligence-Based Medicine","topic":"Balance, Gait, and Falls Prevention","field":"Health Professions","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; McMaster University","funders":"National Institute for Health Research Applied Research Collaboration West; National Institute for Health and Care Research; Northumbria University","keywords":"Inertial measurement unit; Computer science; Terrain; Artificial intelligence; Convolutional neural network; Wearable computer; Context (archaeology); Classifier (UML); Deep learning; Machine learning; Cartography; Geography","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":[0.003016819,0.000198194,0.0003265797,0.000181611,0.0006166518,0.00001305547,0.000293701,0.0001748652,0.0000544563],"category_scores_gemma":[0.001789425,0.0001699717,0.00002655957,0.0001833689,0.0001939409,0.000157334,0.0002339125,0.0004503108,0.00002048227],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009778174,"about_ca_system_score_gemma":0.00008073745,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002164523,"about_ca_topic_score_gemma":0.0002429244,"domain_scores_codex":[0.9975949,0.0004454535,0.0006204838,0.0006053378,0.0002873699,0.000446441],"domain_scores_gemma":[0.9963204,0.002507481,0.0004030344,0.0005496205,0.00005814126,0.0001612801],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0001386253,0.0001601218,0.1251942,0.001395584,0.0001049151,0.00000511957,0.01841106,0.00004790416,0.02398337,0.000593193,0.002135808,0.8278301],"study_design_scores_gemma":[0.001885195,0.0007054769,0.5363505,0.002688106,0.0002272463,0.000001467413,0.04125268,0.400936,0.0000537963,0.0028247,0.01266778,0.0004070299],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2637808,0.0008952057,0.7289913,0.002936557,0.0005583587,0.001362072,0.00006226851,0.0002115724,0.001201866],"genre_scores_gemma":[0.986474,0.004538345,0.006792831,0.0003067668,0.0003829077,0.0002747994,0.0006273066,0.00003469126,0.0005683214],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8274231,"threshold_uncertainty_score":0.6931242,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1161420112987932,"score_gpt":0.406787721307336,"score_spread":0.2906457100085428,"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."}}