arXiv subjects

Matthew S. Mayo

Publications and source records attributed to Matthew S. Mayo.

2 recordsLinked to original sources

arXiv2026

Sample size calculation comparing two survival curves in the presence of nonproportional hazards utilizing the concept of Weighted Relative Time using two different Weibull distributions

Sample size calculations in two-arm clinical trials with survival endpoint often utilize the proportional hazards (PH) assumption. In case of nonproportional hazards (NPH), much research is ongoing with some newer methods developed that work well in specific scenarios. Recent developments have considered the nonparametric Max-Combo logrank test for both design and analysis of survival data with NPH. Although many approaches have been proposed for analyzing time-to-event data with NPH such as Restricted Mean Survival Time, Average Hazard Ratios, and piecewise exponential model with hazard ratios changing over time, translation into sample size calculation is not obvious and relies on simulations for very specific scenarios of NPH. In the parametric setting, accelerated failure time models that allow NPH but require a constant time ratio assumption (PT assumption) have been proposed. Recently, a parametric approach that simultaneously allows both nonproportional hazards (NPH) and nonproportional time (NPT) has been proposed. It utilizes the concept of Relative Time (RT) with two different Weibull distributions for the two trial arms in the case of monotonically increasing/decreasing time ratios. A key limitation is that the hypotheses leading to the sample size calculation rely on a single effect size defined at a specific survival percentile. In our current work, we extend the parent RT method by utilizing the concept of Weighted Relative Time (WRT) that allows us comparison of the full survival curves. Performance of the proposed method is compared to the Max-Combo method under different weighting schemes, and its operation characteristics are studied via simulations.

stat.ME↗

arXiv2018

Relevant Word Order Vectorization for Improved Natural Language Processing in Electronic Healthcare Records

Objective: Electronic health records (EHR) represent a rich resource for conducting observational studies, supporting clinical trials, and more. However, much of the relevant information is stored in an unstructured format that makes it difficult to use. Natural language processing approaches that attempt to automatically classify the data depend on vectorization algorithms that impose structure on the text, but these algorithms were not designed for the unique characteristics of EHR. Here, we propose a new algorithm for structuring so-called free-text that may help researchers make better use of EHR. We call this method Relevant Word Order Vectorization (RWOV). Materials and Methods: As a proof-of-concept, we attempted to classify the hormone receptor status of breast cancer patients treated at the University of Kansas Medical Center during a recent year, from the unstructured text of pathology reports. Our approach attempts to account for the semi-structured way that healthcare providers often enter information. We compared this approach to the ngrams and word2vec methods. Results: Our approach resulted in the most consistently high accuracy, as measured by F1 score and area under the receiver operating characteristic curve (AUC). Discussion: Our results suggest that methods of structuring free text that take into account its context may show better performance, and that our approach is promising. Conclusion: By using a method that accounts for the fact that healthcare providers tend to use certain key words repetitively and that the order of these key words is important, we showed improved performance over methods that do not.

cs.CL↗