{"id":"W2809070429","doi":"10.2196/medinform.7096","title":"Contextual Anonymization for Secondary Use of Big Data in Biomedical Research: Proposal for an Anonymization Matrix","year":2018,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Ethics in Clinical Research","field":"Medicine","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Data anonymization; Big data; Data science; Data mining; Computer security; Information privacy","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01596043,0.0001729904,0.0005485549,0.0006448274,0.0001490729,0.00007342798,0.0009003992,0.001099057,0.0002226332],"category_scores_gemma":[0.0765589,0.0001387836,0.00007311073,0.000908254,0.002393869,0.0008120917,0.0006037997,0.001678599,0.00002183573],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001177903,"about_ca_system_score_gemma":0.006424729,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005136228,"about_ca_topic_score_gemma":0.0006996196,"domain_scores_codex":[0.9935915,0.0002314938,0.001968597,0.0003391562,0.003197164,0.0006720587],"domain_scores_gemma":[0.9838156,0.01097539,0.0003071367,0.001330115,0.002801641,0.0007701857],"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":[0.01324151,0.007640763,0.008214829,0.0224248,0.0003734664,0.00003439361,0.02177304,0.000006544671,0.001608906,0.1278085,0.1942365,0.6026367],"study_design_scores_gemma":[0.02123291,0.01812855,0.002090736,0.002976112,0.00008957749,0.00005757696,0.005901092,0.7422101,0.001434319,0.02995491,0.1753687,0.0005554198],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3935212,0.000119176,0.5583025,0.02049435,0.002074108,0.0196034,0.002640735,0.0002879505,0.002956535],"genre_scores_gemma":[0.6931415,0.0004359559,0.2839721,0.003317561,0.00445185,0.0008558546,0.01189641,0.0001528531,0.001775885],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7422036,"threshold_uncertainty_score":0.9992079,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.674555123165151,"score_gpt":0.6211667524373141,"score_spread":0.05338837072783686,"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."}}