arXiv subjects

Michael P. Wallace

Publications and source records attributed to Michael P. Wallace.

3 recordsLinked to original sources

arXiv2026

Generative AI use in Statistical Research: A Literature Review and Code Generation Case Study

Generative artificial intelligence (GenAI) is a large language model (LLM) that has the ability to generate media based on user-provided prompts. Given the demonstrated capabilities of models such as ChatGPT in information synthesis and programming, there is growing interest in their potential role within the research process. However, little work has evaluated recent GenAI models for research tasks in the domain of statistical research. This case study examines GenAI as a tool for developing a literature review and translating methodology from academic papers into code, for the topic of dynamic treatment regime (DTR) estimation via the dynamic weighted ordinary least squares (dWOLS) approach. Specifically, we utilize ChatGPT-5 and ScholarAI (Sept-Nov 2025 release) in the processes of identifying relevant sources for the literature review, creating summaries of papers, identifying gaps in research, and R code generation to implement methodology. Our findings show that current GenAI models lack the depth and contextual understanding required to accomplish these tasks without careful prompting and supervision of a knowledgeable researcher. Nonetheless, GenAI has potential to increase efficiency of tasks which take advantage of its search and summarization abilities, as well as basic code debugging and algorithm formation. We demonstrate that under a knowledgeable guide, GenAI can function as a research tool, but not as a substitute for methodological expertise.

stat.OT↗

arXiv2025

Photon Absorption Remote Sensing Virtual Histopathology: A Preliminary Exploration of Diagnostic Equivalence to Gold-Standard H&E Staining in Skin Cancer Excisional Biopsies

Photon Absorption Remote Sensing (PARS) enables label-free imaging of subcellular morphology by observing biomolecule specific absorption interactions. Coupled with deep-learning, PARS produces label-free virtual Hematoxylin and Eosin (H&E) stained images in unprocessed tissues. This study evaluates the diagnostic performance of PARS virtual H&E images in excisional skin biopsies, including Squamous (SCC), Basal (BCC) Cell Carcinoma, and normal skin. Sixteen unstained formalin-fixed paraffin-embedded skin excisions were PARS imaged, virtually H&E stained, then chemically stained and imaged at 40x. Seven fellowship trained dermatopathologists assessed all images. Example PARS and chemical H&E whole-slide images from this study are available at the BioImage Archive (https://doi.org/10.6019/S-BIAD2324). Concordance analysis indicates 95.5% agreement between primary diagnoses from PARS versus H&E images (Cohen's k=0.93). Inter-rater reliability was near-perfect for both image types (Fleiss' k=0.89 for PARS, k=0.80 for H&E). For subtype classification, agreement was near-perfect 91% (k=0.73) for SCC and was perfect for BCC. For malignancy confinement (e.g., cancer margins), agreement was 92% between PARS and H&E (k=0.718). During assessment dermatopathologists could not reliably distinguish image origin (PARS vs. H&E), and diagnostic confidence was equivalent. Inter-rater reliability for PARS virtual H&E was consistent with reported histologic evaluation benchmarks. These results indicate that PARS virtual histology may be diagnostically equivalent to chemical H&E staining in dermatopathology diagnostics, while enabling assessment directly from unlabeled slides. In turn, the label-free PARS virtual H&E imaging workflow may preserve tissue for downstream analysis while producing data well-suited for AI integration potentially accelerating and enhancing skin cancer diagnostics.

q-bio.QM↗

arXiv2024

Optimal Dynamic Treatment Regime Estimation in the Presence of Nonadherence

Dynamic treatment regimes (DTRs) are sequences of functions that formalize the process of precision medicine. DTRs take as input patient information and output treatment recommendations. A major focus of the DTR literature has been on the estimation of optimal DTRs, the sequences of decision rules that result in the best outcome in expectation, across the complete population were they to be applied. While there is a rich literature on optimal DTR estimation, to date there has been minimal consideration of the impacts of nonadherence on these estimation procedures. Nonadherence refers to any process through that an individual's prescribed treatment does not match their true treatment. We explore the impacts of nonadherence and demonstrate that generally, when nonadherence is ignored, suboptimal regimes will be estimated. In light of these findings we propose a method for estimating optimal DTRs in the presence of nonadherence. The resulting estimators are consistent and asymptotically normal, with a double robustness property. Using simulations we demonstrate the reliability of these results, and illustrate comparable performance between the proposed estimation procedure adjusting for the impacts of nonadherence and estimators that are computed on data without nonadherence.

stat.ME↗