Selected work

Representative Papers

Finite-sample correction for the covariate-adjusted log-rank test

Aug 11, 2026

In small-sample settings, covariate-adjusted log-rank tests are prone to inflated Type I error rates due to treatment allocation imbalance or a relatively large number of covariates. To address this issue, this work proposes a finite-sample correction method that explicitly accounts for the loss of residual degrees of freedom and the uncertainty in regression coefficient estimation by adjusting the denominator of the test statistic. This approach introduces, for the first time in covariate-adjusted log-rank testing, a correction mechanism that jointly incorporates both degrees-of-freedom reduction and parameter estimation uncertainty. Monte Carlo simulations demonstrate that the proposed correction substantially reduces Type I error rates across a range of scenarios, thereby enhancing the reliability of hypothesis testing in small samples.

0 citationsRead paper

Testing for subgroup treatment effect consistency in the Cox model

Aug 03, 2026

This study addresses the limitation of conventional interaction tests in assessing whether treatment effect differences across subgroups are clinically negligible, which hinders reliable extrapolation of overall efficacy to specific subpopulations. Within the Cox proportional hazards framework, the authors reformulate subgroup treatment effect consistency as an equivalence testing problem, developing methods based on (weighted) treatment–subgroup interaction coefficients to evaluate consistency both between complementary subgroups and between each subgroup and the overall population. They propose a novel normal-parameter optimal equivalence test that maintains asymptotic validity while substantially improving statistical power. Simulation studies demonstrate superior performance over traditional two-one-sided-tests (TOST), and the method is successfully applied to real data from the CANTOS cardiovascular outcomes trial.

0 citationsRead paper

SCTA: An Agentic Framework for Stable and Interpretable Target Gene Discovery from Single-Cell RNA Sequencing

Jul 26, 2026

This study addresses the instability and poor interpretability often encountered in identifying therapeutic target genes from single-cell RNA sequencing data, which stem from the sensitivity of analytical pipelines. To overcome these limitations, the authors propose SCTA, a novel framework that introduces a decision-aware multi-agent orchestration mechanism. SCTA decomposes target gene discovery into specialized agents aligned with critical analytical decisions and constrains their reasoning with structured biological evidence. By integrating differential expression analysis, cell subpopulation selection strategies, and multi-source biological knowledge, the method substantially enhances the stability and interpretability of target gene prioritization. Validation in hereditary chronic pancreatitis demonstrates that SCTA not only improves reproducibility but also successfully recapitulates known disease mechanisms.

0 citationsRead paper

Pseudo-value Based Mean Cumulative Count Regression

Jun 22, 2026

This study addresses recurrent event data subject to right censoring and a terminal event by proposing a pseudo-value–based regression approach to model the effects of covariates on the mean cumulative function (MCF) and its area under the curve (AUMCF) at fixed time points. The method systematically extends the pseudo-value regression framework to estimate covariate effects on both MCF and AUMCF, constructing pseudo-observations via influence functions and enabling efficient inference through generalized estimating equations or ordinary least squares. Computationally straightforward and interpretable, the approach is compatible with standard regression software. Simulation studies demonstrate its favorable performance across diverse recurrent event settings, exhibiting accurate estimation, proper confidence interval coverage, controlled Type I error rates, and high statistical power. The method is successfully applied to the ORATORIO clinical trial data.

0 citationsRead paper

Introducing precision-weighted bias as a performance measure to inform the inclusion of adaptive designs in meta-analysis

Jun 10, 2026

Current systematic review guidelines restrict the inclusion of adaptive designs due to concerns about bias, yet they overlook the joint influence of bias and information content. This work proposes “precision-weighted bias” as a novel metric, which accounts for each study’s unconditional bias scaled by its precision, thereby more accurately reflecting its contribution to overall meta-analytic bias. Theoretical derivation demonstrates that the total bias in a meta-analysis is in fact a precision-weighted average of individual study biases, rather than a simple arithmetic mean. Simulation studies further reveal that although adaptive designs may exhibit unweighted bias, their precision-weighted bias is often negligible—resulting in minimal impact on pooled estimates when included. These findings provide both theoretical justification and methodological support for the appropriate inclusion of adaptive designs in systematic reviews.

0 citationsRead paper

Recent publications

Latest Papers

Finite-sample correction for the covariate-adjusted log-rank test

Aug 11, 2026

In small-sample settings, covariate-adjusted log-rank tests are prone to inflated Type I error rates due to treatment allocation imbalance or a relatively large number of covariates. To address this issue, this work proposes a finite-sample correction method that explicitly accounts for the loss of residual degrees of freedom and the uncertainty in regression coefficient estimation by adjusting the denominator of the test statistic. This approach introduces, for the first time in covariate-adjusted log-rank testing, a correction mechanism that jointly incorporates both degrees-of-freedom reduction and parameter estimation uncertainty. Monte Carlo simulations demonstrate that the proposed correction substantially reduces Type I error rates across a range of scenarios, thereby enhancing the reliability of hypothesis testing in small samples.

0 citationsRead paper

Testing for subgroup treatment effect consistency in the Cox model

Aug 03, 2026

This study addresses the limitation of conventional interaction tests in assessing whether treatment effect differences across subgroups are clinically negligible, which hinders reliable extrapolation of overall efficacy to specific subpopulations. Within the Cox proportional hazards framework, the authors reformulate subgroup treatment effect consistency as an equivalence testing problem, developing methods based on (weighted) treatment–subgroup interaction coefficients to evaluate consistency both between complementary subgroups and between each subgroup and the overall population. They propose a novel normal-parameter optimal equivalence test that maintains asymptotic validity while substantially improving statistical power. Simulation studies demonstrate superior performance over traditional two-one-sided-tests (TOST), and the method is successfully applied to real data from the CANTOS cardiovascular outcomes trial.

0 citationsRead paper

SCTA: An Agentic Framework for Stable and Interpretable Target Gene Discovery from Single-Cell RNA Sequencing

Jul 26, 2026

This study addresses the instability and poor interpretability often encountered in identifying therapeutic target genes from single-cell RNA sequencing data, which stem from the sensitivity of analytical pipelines. To overcome these limitations, the authors propose SCTA, a novel framework that introduces a decision-aware multi-agent orchestration mechanism. SCTA decomposes target gene discovery into specialized agents aligned with critical analytical decisions and constrains their reasoning with structured biological evidence. By integrating differential expression analysis, cell subpopulation selection strategies, and multi-source biological knowledge, the method substantially enhances the stability and interpretability of target gene prioritization. Validation in hereditary chronic pancreatitis demonstrates that SCTA not only improves reproducibility but also successfully recapitulates known disease mechanisms.

0 citationsRead paper

Pseudo-value Based Mean Cumulative Count Regression

Jun 22, 2026

This study addresses recurrent event data subject to right censoring and a terminal event by proposing a pseudo-value–based regression approach to model the effects of covariates on the mean cumulative function (MCF) and its area under the curve (AUMCF) at fixed time points. The method systematically extends the pseudo-value regression framework to estimate covariate effects on both MCF and AUMCF, constructing pseudo-observations via influence functions and enabling efficient inference through generalized estimating equations or ordinary least squares. Computationally straightforward and interpretable, the approach is compatible with standard regression software. Simulation studies demonstrate its favorable performance across diverse recurrent event settings, exhibiting accurate estimation, proper confidence interval coverage, controlled Type I error rates, and high statistical power. The method is successfully applied to the ORATORIO clinical trial data.

0 citationsRead paper

Introducing precision-weighted bias as a performance measure to inform the inclusion of adaptive designs in meta-analysis

Jun 10, 2026

Current systematic review guidelines restrict the inclusion of adaptive designs due to concerns about bias, yet they overlook the joint influence of bias and information content. This work proposes “precision-weighted bias” as a novel metric, which accounts for each study’s unconditional bias scaled by its precision, thereby more accurately reflecting its contribution to overall meta-analytic bias. Theoretical derivation demonstrates that the total bias in a meta-analysis is in fact a precision-weighted average of individual study biases, rather than a simple arithmetic mean. Simulation studies further reveal that although adaptive designs may exhibit unweighted bias, their precision-weighted bias is often negligible—resulting in minimal impact on pooled estimates when included. These findings provide both theoretical justification and methodological support for the appropriate inclusion of adaptive designs in systematic reviews.

0 citationsRead paper