arXiv subjects

Marie Hudson

Publications and source records attributed to Marie Hudson.

5 recordsLinked to original sources

arXiv2026

DoctorAgents: an agentic framework to iteratively refine AutoML pipeline for small clinical temporal data

Clinical machine learning (ML) has the potential to support high-stakes medical decision-making, but reliable deployment is often constrained by scarce, heterogeneous, and temporal complexity. Developing effective ML pipelines for such data remains time-consuming and error-prone, while existing automated machine learning (AutoML) systems only partially address this challenge because they largely rely on brute-force search over predefined spaces and lack explicit reasoning and memory. We therefore reformulate AutoML for small clinical data from exhaustive search to reasoning-driven refinement. We propose DoctorAgents, an agentic AI framework that autonomously constructs and optimizes end-to-end ML pipelines through specialized large language model (LLM) agents for generation, validation, and refinement. DoctorAgents backpropagates natural-language feedback through textual gradient descent to perform targeted updates without exhaustive search. Experiments across diverse clinical tasks show that DoctorAgents consistently outperforms established AutoML baselines while producing more interpretable task-specific representations.

cs.AI↗

arXiv2025

The efficiencies of pilot feasibility trials in rare diseases using Bayesian methods

Pilot feasibility studies play a pivotal role in the development of clinical trials for rare diseases, where small populations and slow recruitment often threaten trial viability. While such studies are commonly used to assess operational parameters, they also offer a valuable opportunity to inform the design and analysis of subsequent definitive trials-particularly through the use of Bayesian methods. In this paper, we demonstrate how data from a single, protocol-aligned pilot study can be incorporated into a definitive trial using robust meta-analytic-predictive priors. We focus on the case of a binary efficacy outcome, motivated by a feasibility trial of intravenous immunoglobulin tapering in autoimmune inflammatory myopathies. Through simulation studies, we evaluate the operating characteristics of trials informed by pilot data, including sample size, expected trial duration, and the probability of meeting recruitment targets. Our findings highlight the operational and ethical advantages of leveraging pilot data via robust Bayesian priors, and offer practical guidance for their application in rare disease settings.

stat.AP↗

arXiv2025

Multivariate regression with missing response data for modelling regional DNA methylation QTLs

Identifying genetic regulators of DNA methylation (mQTLs) with multivariate models enhances statistical power, but is challenged by missing data from bisulfite sequencing. Standard imputation-based methods can introduce bias, limiting reliable inference. We propose \texttt{missoNet}, a novel convex estimation framework that jointly estimates regression coefficients and the precision matrix from data with missing responses. By using unbiased surrogate estimators, our three-stage procedure avoids imputation while simultaneously performing variable selection and learning the conditional dependence structure among responses. We establish theoretical error bounds, and our simulations demonstrate that \texttt{missoNet} consistently outperforms existing methods in both prediction and sparsity recovery. In a real-world mQTL analysis of the CARTaGENE cohort, \texttt{missoNet} achieved superior predictive accuracy and false-discovery control on a held-out validation set, identifying known and credible novel genetic associations. The method offers a robust, efficient, and theoretically grounded tool for genomic analyses, and is available as an R package.

stat.ME↗

arXiv2024

An adaptive enrichment design using Bayesian model averaging for selection and threshold-identification of tailoring variables

Precision medicine stands as a transformative approach in healthcare, offering tailored treatments that can enhance patient outcomes and reduce healthcare costs. As understanding of complex disease improves, clinical trials are being designed to detect subgroups of patients with enhanced treatment effects. Biomarker-driven adaptive enrichment designs, which enroll a general population initially and later restrict accrual to treatment-sensitive patients, are gaining popularity. Current practice often assumes either pre-trial knowledge of biomarkers defining treatment-sensitive subpopulations or a simple, linear relationship between continuous markers and treatment effectiveness. Motivated by a trial studying rheumatoid arthritis treatment, we propose a Bayesian adaptive enrichment design which identifies important tailoring variables out of a larger set of candidate biomarkers. Our proposed design is equipped with a flexible modelling framework where the effects of continuous biomarkers are introduced using free knot B-splines. The parameters of interest are then estimated by marginalizing over the space of all possible variable combinations using Bayesian model averaging. At interim analyses, we assess whether a biomarker-defined subgroup has enhanced or reduced treatment effects, allowing for early termination due to efficacy or futility and restricting future enrollment to treatment-sensitive patients. We consider pre-categorized and continuous biomarkers, the latter of which may have complex, nonlinear relationships to the outcome and treatment effect. Using simulations, we derive the operating characteristics of our design and compare its performance to two existing approaches.

stat.ME↗

arXiv2021

Detecting differentially methylated regions in bisulfite sequencing data using quasi-binomial mixed models with smooth covariate effect estimates

Identifying disease-associated changes in DNA methylation can help to gain a better understanding of disease etiology. Bisulfite sequencing technology allows the generation of methylation profiles at single base of DNA. We previously developed a method for estimating smooth covariate effects and identifying differentially methylated regions (DMRs) from bisulfite sequencing data, which copes with experimental errors and variable read depths; this method utilizes the binomial distribution to characterize the variability in the methylated counts. However, bisulfite sequencing data frequently include low-count integers and can exhibit over or under dispersion relative to the binomial distribution. We present a substantial improvement to our previous work by proposing a quasi-likelihood-based regional testing approach which accounts for multiplicative and additive sources of dispersion. We demonstrate the theoretical properties of the resulting tests, as well as their marginal and conditional interpretations. Simulations show that the proposed method provides correct inference for smooth covariate effects and captures the major methylation patterns with excellent power.

stat.ME↗