Bayesian Diagnostic Test Evaluation — bayesiandiagnostic • ClinicoPath

Comprehensive Bayesian evaluation of diagnostic test performance including sensitivity, specificity, predictive values, likelihood ratios, and diagnostic odds ratios with full uncertainty quantification. Supports meta-analysis, hierarchical modeling, and comparative test evaluation.

Usage

bayesiandiagnostic(
  data,
  test_results,
  gold_standard,
  test_positive_level,
  disease_positive_level,
  study_id = NULL,
  patient_id = NULL,
  covariates,
  analysis_type = "single_test",
  comparison_test = NULL,
  comparison_positive_level,
  prior_type = "informative",
  prior_sensitivity_mean = 0.8,
  prior_sensitivity_precision = 10,
  prior_specificity_mean = 0.9,
  prior_specificity_precision = 10,
  correlation_prior = 0,
  bivariate_model = TRUE,
  hierarchical_model = FALSE,
  covariate_effects = FALSE,
  heterogeneity_model = "random",
  mcmc_samples = 10000,
  mcmc_burnin = 5000,
  mcmc_thin = 1,
  mcmc_chains = 3,
  mcmc_adapt = TRUE,
  convergence_criterion = 1.1,
  compute_sensitivity = TRUE,
  compute_specificity = TRUE,
  compute_ppv = TRUE,
  compute_npv = TRUE,
  compute_likelihood_ratios = TRUE,
  compute_diagnostic_odds_ratio = TRUE,
  compute_accuracy = TRUE,
  compute_auc = FALSE,
  prevalence_prior = 0.5,
  decision_analysis = FALSE,
  cost_fn = 1,
  cost_fp = 1,
  utility_tp = 1,
  utility_tn = 1,
  threshold_analysis = FALSE,
  threshold_metric = "youden",
  show_summary_statistics = TRUE,
  show_posterior_distributions = TRUE,
  show_credible_intervals = TRUE,
  show_convergence_diagnostics = TRUE,
  show_model_comparison = FALSE,
  show_predictive_checks = FALSE,
  show_clinical_interpretation = TRUE,
  credible_interval = 0.95,
  hpdi = TRUE,
  set_seed = TRUE,
  seed_value = 42,
  parallel_chains = TRUE,
  n_cores = 2,
  robust_estimation = FALSE,
  outlier_detection = FALSE,
  publication_bias = FALSE,
  sensitivity_analysis = FALSE
)

Arguments

data: The data as a data frame for Bayesian diagnostic test evaluation.
test_results: .
gold_standard: .
test_positive_level: .
disease_positive_level: .
study_id: .
patient_id: .
covariates: .
analysis_type: .
comparison_test: .
comparison_positive_level: .
prior_type: .
prior_sensitivity_mean: .
prior_sensitivity_precision: .
prior_specificity_mean: .
prior_specificity_precision: .
correlation_prior: .
bivariate_model: .
hierarchical_model: .
covariate_effects: .
heterogeneity_model: .
mcmc_samples: .
mcmc_burnin: .
mcmc_thin: .
mcmc_chains: .
mcmc_adapt: .
convergence_criterion: .
compute_sensitivity: .
compute_specificity: .
compute_ppv: .
compute_npv: .
compute_likelihood_ratios: .
compute_diagnostic_odds_ratio: .
compute_accuracy: .
compute_auc: .
prevalence_prior: .
decision_analysis: .
cost_fn: .
cost_fp: .
utility_tp: .
utility_tn: .
threshold_analysis: .
threshold_metric: .
show_summary_statistics: .
show_posterior_distributions: .
show_credible_intervals: .
show_convergence_diagnostics: .
show_model_comparison: .
show_predictive_checks: .
show_clinical_interpretation: .
credible_interval: .
hpdi: .
set_seed: .
seed_value: .
parallel_chains: .
n_cores: .
robust_estimation: .
outlier_detection: .
publication_bias: .
sensitivity_analysis: .

Value

A results object containing:

`results$summaryStatistics`					a table
`results$diagnosticPerformance`					a table
`results$sensitivitySpecificity`					a table
`results$predictiveValues`					a table
`results$likelihoodRatios`					a table
`results$diagnosticOddsRatio`					a table
`results$convergenceDiagnostics`					a table
`results$posteriorSummary`					a table
`results$modelComparison`					a table
`results$metaAnalysisResults`					a table
`results$heterogeneityAssessment`					a table
`results$covariateEffects`					a table
`results$thresholdAnalysis`					a table
`results$decisionAnalysis`					a table
`results$posteriorPredictiveChecks`					a table
`results$sensitivityAnalysis`					a table
`results$clinicalInterpretation`					a html
`results$methodsExplanation`					a html
`results$rocCurve`					an image
`results$posteriorDistributionsPlot`					an image
`results$sensitivitySpecificityPlot`					an image
`results$predictiveValuesPlot`					an image
`results$likelihoodRatioPlot`					an image
`results$convergencePlots`					an image
`results$forestPlot`					an image
`results$heterogeneityPlot`					an image
`results$thresholdOptimizationPlot`					an image
`results$decisionCurvePlot`					an image

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$summaryStatistics$asDF

as.data.frame(results$summaryStatistics)

Examples

# Bayesian diagnostic test evaluation
bayesiandiagnostic(
    data = diagnostic_data,
    test_results = "test_outcome",
    gold_standard = "true_diagnosis",
    prior_sensitivity_mean = 0.8,
    prior_specificity_mean = 0.9
)