Browsing by Author "Miller, Dylan"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Introduction to applied statistics: open textbook series in statistics(2024) Su, Wanhua; Miller, Dylan; Mewhort, Clarissa; Chipman, Hugh; Fedoruk, JohnThis book aims to provide students taking the first course in introductory statistics with open learning materials to master basic statistical concepts and techniques and to give demonstrations on conducting fundamental statistical analysis using the free statistical software R Commander. Each chapter generally includes a statement of learning outcomes, course notes, review exercises, self-assessment quiz, and homework assignment questions. The book is based on instructor course notes for STAT 151 (Introduction to Applied Statistics) at MacEwan University. In December 2015, the online version of STAT 151, including module notes, quizzes, homework assignment questions and marking rubrics, and a lab manual in R Commander was developed, leading to the creation of this textbook. Each homework assignment has two parts; students must complete Part A by hand and Part B with R Commander. Most data sets for the assignment, assignment questions, and quiz questions are adapted from popular introductory statistics textbooks such as Introductory Statistics by Neil Weiss and Intro STATS by Richard D. De Veaux, Paul F. Velleman, David E. Bock, and Paul D. Velleman. The online STAT 151 was completed and offered for the first time in Spring 2018. This open textbook is the revised and enriched version of that online course. The only prerequisite of this book is high school mathematics; most students take STAT 151 in the first year of their post-secondary education. R Commander is taught instead of R/RStudio as the software for the lab component to avoid focusing on the programming component needed for R/R Studio. In a future edition, there are plans to include a lab manual with command lines in R/RStudio. This book introduces one-sided confidence intervals to help students understand the computer output of hypothesis testing in R Commander.Item A UNet pipeline for segmentation of new MS lesions(2021) Efird, Cory; Miller, Dylan; Cobzas, DanaA pipeline for the second multiple sclerosis segmentation challenge (MSSEG-2) hosted by MICCAI is proposed. Two FLAIR images taken at different time-points are used as a multi-channel input to a 3D CNN to detect new lesions. Patch sampling strategies are adopted to keep the input volume shape manageable in terms of memory requirements. To further improve results, multiple models and patch orientations are ensembled. Performance is evaluated against nn-UNet.