# Extract operating characteristics data
extract_oc_data <- function(boinet_result) {
  dose_levels <- names(boinet_result$n.patient)
  
  data.frame(
    dose_level = dose_levels,
    toxicity_prob = as.numeric(boinet_result$toxprob),
    efficacy_prob = as.numeric(boinet_result$effprob),
    n_patients = as.numeric(boinet_result$n.patient),
    selection_prob = as.numeric(boinet_result$prop.select),
    stringsAsFactors = FALSE
  )
}

# Create summary statistics
create_summary_stats <- function(boinet_result) {
  oc_data <- extract_oc_data(boinet_result)
  optimal_dose <- oc_data$dose_level[which.max(oc_data$selection_prob)]
  
  list(
    optimal_dose = optimal_dose,
    max_selection_prob = max(oc_data$selection_prob),
    early_stop_rate = as.numeric(boinet_result$prop.stop),
    avg_duration = as.numeric(boinet_result$duration),
    total_patients = sum(oc_data$n_patients),
    design_type = class(boinet_result)[1]
  )
}

# Create formatted text summaries
create_text_summary <- function(boinet_result) {
  stats <- create_summary_stats(boinet_result)
  
  sprintf(
    "The %s design selected dose level %s in %.1f%% of trials, with an average trial duration of %.0f days and early stopping rate of %.1f%%.",
    toupper(gsub("\\.", "-", stats$design_type)),
    stats$optimal_dose,
    stats$max_selection_prob,
    stats$avg_duration,
    stats$early_stop_rate
  )
}

# Create design parameters table
create_design_table <- function(result) {
  design_params <- data.frame(
    Parameter = c("Target Toxicity Rate (φ)", "Target Efficacy Rate (δ)", 
                  "Lower Toxicity Boundary (λ₁)", "Upper Toxicity Boundary (λ₂)",
                  "Efficacy Boundary (η₁)", "Early Stop Rate (%)", 
                  "Average Duration (days)", "Toxicity Assessment Window (days)",
                  "Efficacy Assessment Window (days)", "Accrual Rate (days)"),
    Value = c(result$phi, result$delta, result$lambda1, result$lambda2,
              result$eta1, result$prop.stop, result$duration,
              result$tau.T, result$tau.E, result$accrual),
    stringsAsFactors = FALSE
  )
  
  # Format values
  design_params$Value <- round(as.numeric(design_params$Value), 3)
  
  if (gt_available) {
    design_params %>%
      gt() %>%
      tab_header(
        title = "TITE-BOIN-ET Design Parameters",
        subtitle = paste("Based on", result$n.sim, "simulated trials")
      ) %>%
      fmt_number(columns = "Value", decimals = 3) %>%
      cols_align(align = "left", columns = "Parameter") %>%
      cols_align(align = "center", columns = "Value")
  } else {
    # Fallback to basic table
    print(design_params)
  }
}

# Create operating characteristics table
create_oc_table <- function(result) {
  dose_levels <- names(result$n.patient)
  
  oc_data <- data.frame(
    `Dose Level` = dose_levels,
    `True Toxicity Probability` = round(as.numeric(result$toxprob), 3),
    `True Efficacy Probability` = round(as.numeric(result$effprob), 3),
    `Average N Treated` = round(as.numeric(result$n.patient), 1),
    `Selection Probability (%)` = round(as.numeric(result$prop.select), 1),
    check.names = FALSE
  )
  
  if (gt_available) {
    oc_data %>%
      gt() %>%
      tab_header(
        title = "Operating Characteristics",
        subtitle = "TITE-BOIN-ET Design Simulation Results"
      ) %>%
      fmt_number(columns = c("True Toxicity Probability", "True Efficacy Probability"), decimals = 3) %>%
      fmt_number(columns = "Average N Treated", decimals = 1) %>%
      fmt_number(columns = "Selection Probability (%)", decimals = 1) %>%
      cols_align(align = "center", columns = everything()) %>%
      cols_align(align = "left", columns = "Dose Level") %>%
      tab_style(
        style = cell_fill(color = "lightblue"),
        locations = cells_body(
          rows = `Selection Probability (%)` == max(`Selection Probability (%)`)
        )
      )
  } else {
    # Fallback to basic table
    print(oc_data)
  }
}

# Create mock result for demonstration
boinet_result <- list(
  toxprob = c("1" = 0.02, "2" = 0.08, "3" = 0.15, "4" = 0.25, "5" = 0.40),
  effprob = c("1" = 0.10, "2" = 0.20, "3" = 0.35, "4" = 0.50, "5" = 0.65),
  n.patient = c("1" = 8.2, "2" = 12.5, "3" = 15.8, "4" = 10.3, "5" = 7.2),
  prop.select = c("1" = 5.2, "2" = 18.7, "3" = 42.1, "4" = 28.3, "5" = 5.7),
  phi = 0.30, delta = 0.60, lambda1 = 0.03, lambda2 = 0.42, eta1 = 0.36,
  tau.T = 28, tau.E = 42, accrual = 7, duration = 156.3, prop.stop = 3.2, n.sim = 1000
)
class(boinet_result) <- "tite.boinet"

if (ggplot2_available) {
  # Create data frame for plotting
  plot_data <- data.frame(
    dose_level = names(boinet_result$n.patient),
    selection_prob = as.numeric(boinet_result$prop.select)
  )
  
  plot_data %>%
    ggplot(aes(x = dose_level, y = selection_prob)) +
    geom_col(fill = "steelblue", alpha = 0.7) +
    geom_text(aes(label = paste0(round(selection_prob, 1), "%")), 
              vjust = -0.3) +
    labs(
      x = "Dose Level",
      y = "Selection Probability (%)",
      title = "TITE-BOIN-ET Dose Selection Performance"
    ) +
    theme_minimal()
} else {
  cat("ggplot2 package not available for visualization.\n")
}

if (ggplot2_available) {
  # Create risk-benefit plot data
  rb_data <- data.frame(
    dose_level = names(boinet_result$n.patient),
    toxicity_prob = as.numeric(boinet_result$toxprob),
    efficacy_prob = as.numeric(boinet_result$effprob),
    selection_prob = as.numeric(boinet_result$prop.select)
  )
  
  rb_data %>%
    ggplot(aes(x = toxicity_prob, y = efficacy_prob)) +
    geom_point(aes(size = selection_prob), alpha = 0.7, color = "darkred") +
    geom_text(aes(label = dose_level), vjust = -1.5) +
    scale_size_continuous(name = "Selection\nProbability (%)", range = c(3, 12)) +
    labs(
      x = "True Toxicity Probability",
      y = "True Efficacy Probability",
      title = "Risk-Benefit Profile"
    ) +
    theme_minimal()
} else {
  cat("ggplot2 package not available for risk-benefit visualization.\n")
}

# Demonstrate the concept with executable code
scenarios <- list(
  conservative = list(phi = 0.25, delta = 0.50, name = "Conservative"),
  standard = list(phi = 0.30, delta = 0.60, name = "Standard"),
  aggressive = list(phi = 0.35, delta = 0.70, name = "Aggressive")
)

# Function to create scenario-specific summaries
create_scenario_summary <- function(scenario) {
  sprintf(
    "Scenario: %s (φ=%.2f, δ=%.2f)",
    scenario$name, scenario$phi, scenario$delta
  )
}

# Generate summaries for all scenarios
scenario_summaries <- lapply(scenarios, create_scenario_summary)
cat("Available scenarios:\n")
#> Available scenarios:
for(summary in scenario_summaries) {
  cat("-", summary, "\n")
}
#> - Scenario: Conservative (φ=0.25, δ=0.50) 
#> - Scenario: Standard (φ=0.30, δ=0.60) 
#> - Scenario: Aggressive (φ=0.35, δ=0.70)

# Extract results for conditional logic
oc_data <- extract_oc_data(boinet_result)
max_selection_prob <- max(oc_data$selection_prob)
optimal_dose <- oc_data$dose_level[which.max(oc_data$selection_prob)]

# Conditional content based on results
if (max_selection_prob > 40) {
  recommendation <- "Strong evidence for optimal dose identification"
  confidence_level <- "High"
} else if (max_selection_prob > 25) {
  recommendation <- "Moderate evidence for dose selection"
  confidence_level <- "Moderate"  
} else {
  recommendation <- "Weak evidence - consider design modifications"
  confidence_level <- "Low"
}

cat("Recommendation Confidence:", confidence_level, "\n")
#> Recommendation Confidence: High
cat("Analysis Conclusion:", recommendation, "\n")
#> Analysis Conclusion: Strong evidence for optimal dose identification
cat(sprintf("The optimal dose level %s was selected in %.1f%% of simulations.\n", 
            optimal_dose, max_selection_prob))
#> The optimal dose level 3 was selected in 42.1% of simulations.