Interactive Clinical Model Validation — clinicalvalidationinteractive • ClinicoPath

Interactive clinical prediction model validation with real-time parameter validation, intelligent defaults, and guided workflows. Features automatic parameter correction, clinical scenario presets, and dynamic threshold optimization for medical research and diagnostic applications.

Usage

clinicalvalidationinteractive(
  data,
  outcome,
  outcomeLevel,
  predictors,
  time_variable = NULL,
  clinical_preset = "custom",
  model_type = "logistic",
  validation_method = "bootstrap",
  bootstrap_samples = 1000,
  cv_folds = 10,
  cv_repeats = 3,
  holdout_proportion = 0.25,
  stratified_sampling = TRUE,
  clinical_context = "diagnosis",
  prevalence_adjustment = FALSE,
  population_prevalence = 10,
  cost_matrix = "equal",
  fn_fp_cost_ratio = 2,
  min_sensitivity = 0.8,
  min_specificity = 0.8,
  min_ppv = 0.7,
  min_npv = 0.9,
  auto_optimize_threshold = FALSE,
  optimization_metric = "youden",
  performance_metrics = "all",
  confidence_level = 0.95,
  show_realtime_metrics = TRUE,
  show_parameter_warnings = TRUE,
  show_clinical_guidance = TRUE,
  show_model_summary = TRUE,
  show_performance_table = TRUE,
  show_calibration_plot = TRUE,
  show_roc_curve = TRUE,
  show_threshold_optimization = FALSE,
  show_clinical_interpretation = TRUE,
  missing_data_handling = "complete_cases",
  set_seed = TRUE,
  seed_value = 42
)

Arguments

data: The data as a data frame for clinical model validation.
outcome: .
outcomeLevel: .
predictors: .
time_variable: .
clinical_preset: .
model_type: .
validation_method: .
bootstrap_samples: .
cv_folds: .
cv_repeats: .
holdout_proportion: .
stratified_sampling: .
clinical_context: .
prevalence_adjustment: .
population_prevalence: .
cost_matrix: .
fn_fp_cost_ratio: .
min_sensitivity: .
min_specificity: .
min_ppv: .
min_npv: .
auto_optimize_threshold: .
optimization_metric: .
performance_metrics: .
confidence_level: .
show_realtime_metrics: .
show_parameter_warnings: .
show_clinical_guidance: .
show_model_summary: .
show_performance_table: .
show_calibration_plot: .
show_roc_curve: .
show_threshold_optimization: .
show_clinical_interpretation: .
missing_data_handling: .
set_seed: .
seed_value: .

Value

A results object containing:

`results$modelSummary`					Summary of the fitted model
`results$validationResults`					Comprehensive validation performance metrics with confidence intervals
`results$performanceWarnings`					Real-time warnings and clinical recommendations
`results$prevalenceAnalysis`					Impact of disease prevalence on predictive values
`results$thresholdOptimization`					Optimal decision thresholds for different clinical scenarios
`results$calibrationAssessment`					Calibration performance metrics
`results$clinicalGuidelines`					Evidence-based clinical interpretation and recommendations
`results$rocCurve`					ROC curve showing optimal decision thresholds
`results$calibrationPlot`					Model calibration assessment plot
`results$thresholdPlot`					Performance metrics across decision thresholds
`results$validationCurves`					Learning curves showing model performance vs sample size
`results$realtimeMetrics`					Dynamic performance metric updates
`results$interactiveGuidance`					Dynamic clinical decision guidance
`results$parameterValidation`					Real-time parameter validation and warnings
`results$exportSummary`					Comprehensive validation summary for clinical reporting

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

results$modelSummary$asDF

as.data.frame(results$modelSummary)

Examples

# Interactive clinical validation with presets
clinical_validation_interactive(
    data = clinical_data,
    outcome = "diagnosis",
    predictors = c("biomarker1", "biomarker2", "age"),
    clinical_preset = "diagnostic_biomarker",
    auto_optimize = TRUE
)