Survival Tools

Configuration File: survival_tools.json Tool Type: Local Tools Count: 3

This page contains all tools defined in the survival_tools.json configuration file.

Available Tools

Survival_cox_regression (Type: SurvivalTool)

Fit Cox proportional hazards regression model to assess effect of covariates on survival. Returns…

Survival_cox_regression tool specification

Tool Information:

  • Name: Survival_cox_regression

  • Type: SurvivalTool

  • Description: Fit Cox proportional hazards regression model to assess effect of covariates on survival. Returns hazard ratios with 95% CI, p-values, and significance for each covariate. HR > 1 = increased hazard (worse survival); HR < 1 = decreased hazard (better survival). Use for: multivariate survival analysis, identifying prognostic biomarkers, adjusting for confounders.

Parameters:

  • operation (string) (required) Operation type

  • durations (array) (required) Observed time durations

  • event_observed (array) (required) Event indicators (1=event, 0=censored)

  • covariates (object) (required) Dict mapping covariate name to array of values. E.g., {‘age’: [45, 62, …], ‘stage’: [1, 2, …]}

Example Usage:

query = {
    "name": "Survival_cox_regression",
    "arguments": {
        "operation": "example_value",
        "durations": ["item1", "item2"],
        "event_observed": ["item1", "item2"],
        "covariates": "example_value"
    }
}
result = tu.run(query)

Survival_kaplan_meier (Type: SurvivalTool)

Compute Kaplan-Meier survival estimates from time-to-event data. Returns survival probability at …

Survival_kaplan_meier tool specification

Tool Information:

  • Name: Survival_kaplan_meier

  • Type: SurvivalTool

  • Description: Compute Kaplan-Meier survival estimates from time-to-event data. Returns survival probability at each event time, number at risk, events, censored observations, and median survival time. Supports optional stratification by group for visual comparison. Use for: analyzing patient survival in clinical data, time-to-event analysis, comparing treatment groups.

Parameters:

  • operation (string) (required) Operation type

  • durations (array) (required) Observed time durations (e.g., months to event or censoring). All values must be >= 0.

  • event_observed (array) (required) Event indicator: 1 = event occurred (death/relapse), 0 = censored. Same length as durations.

  • group_labels ([‘array’, ‘null’]) (optional) Optional group labels for stratified KM analysis (e.g., [‘high’, ‘low’, ‘high’, …])

Example Usage:

query = {
    "name": "Survival_kaplan_meier",
    "arguments": {
        "operation": "example_value",
        "durations": ["item1", "item2"],
        "event_observed": ["item1", "item2"]
    }
}
result = tu.run(query)

Survival_log_rank_test (Type: SurvivalTool)

Perform Mantel-Cox log-rank test to compare survival between two groups. Tests null hypothesis th…

Survival_log_rank_test tool specification

Tool Information:

  • Name: Survival_log_rank_test

  • Type: SurvivalTool

  • Description: Perform Mantel-Cox log-rank test to compare survival between two groups. Tests null hypothesis that survival curves are identical. Returns chi-squared statistic, p-value, observed vs expected events per group. A p-value < 0.05 indicates statistically significant difference in survival. Use for: comparing treatment arms, high vs low expression groups.

Parameters:

  • operation (string) (required) Operation type

  • durations_a (array) (required) Time durations for group A

  • events_a (array) (required) Event indicators for group A (1=event, 0=censored)

  • durations_b (array) (required) Time durations for group B

  • events_b (array) (required) Event indicators for group B (1=event, 0=censored)

Example Usage:

query = {
    "name": "Survival_log_rank_test",
    "arguments": {
        "operation": "example_value",
        "durations_a": ["item1", "item2"],
        "events_a": ["item1", "item2"],
        "durations_b": ["item1", "item2"],
        "events_b": ["item1", "item2"]
    }
}
result = tu.run(query)