DATA SCIENCE IN MEDICAL PHYSICS
Concepts and principles of statistics and machine learning for medical physics-related research problems. Topics covered include probability and independence, discrete and continuous random variables and statistical distributions, random sampling and central limit theorem, inference for means, variances, proportions, moment generating functions, maximum likelihood, hypothesis testing, ANOVA, linear regression, correlation and basic design of experiments with application to quality assurance, reliability, and reproducibility.
Topics
- probability and independence
- discrete and continuous random variables and statistical distributions
- random sampling and central limit theorem
- inference for means
- variances
- proportions
- moment generating functions
- maximum likelihood
- hypothesis testing
- ANOVA
- linear regression
- correlation and basic design of experiments with application to quality assurance
- reliability
- and reproducibility
Topics explicitly listed in official catalogue description
Prerequisites & course information
Requisites: ( PHYSICS/B M E/H ONCOL/MED PHYS 501 and B M E/MED PHYS 573 ) or ( STAT/MATH 309 or 431 ) or graduate/professional standing Course Designation: Grad 50% - Counts toward 50% graduate coursework requirement Repeatable for Credit: No Last Taught: Spring 2026 Learning Outcomes: 1. Analyze experimental data by calculating mathematical expectation, co-variance, and variance for both discrete and continuous random variables from experimental measurements and determine statistical parameters using parametric statistical models Audience: Both Grad & Undergrad 2. Draw proper statistical conclusions from experimental data by constructing confidence intervals for estimated statistical parameters from data at a given statistical significance level and formulate a statistical hypothesis for drawing possible conclusions. Construct test statistics to perform hypothesis testing using the P-value method and the rejection-zone method at a given significance level Audience: Both Grad & Undergrad 3. Perform supervised learning using linear regression methods and interpret the regression results for experimental data as well as perform supervised learning using logistic regression methods for classification tasks such as image segmentation Audience: Both Grad & Undergrad 4. Perform unsupervised learning using K-means clustering, Expectation-Maximization (EM) methods, and kernel tricks to recognize underlying patterns in experimental data as well as perform supervised learning using deep neural network modeling and backpropagation learning schemes Audience: Graduate Learning Outcomes: 1. Analyze experimental data by calculating mathematical expectation, co-variance, and variance for both discrete and continuous random variables from experimental measurements and determine statistical parameters using parametric statistical models Audience: Both Grad & Undergrad 2. Draw proper statistical conclusions from experimental data by constructing confidence intervals for estimated statistical parameters from data at a given statistical significance level and formulate a statistical hypothesis for drawing possible conclusions. Construct test statistics to perform hypothesis testing using the P-value method and the rejection-zone method at a given significance level Audience: Both Grad & Undergrad 3. Perform supervised learning using linear regression methods and interpret the regression results for experimental data as well as perform supervised learning using logistic regression methods for classification tasks such as image segmentation Audience: Both Grad & Undergrad 4. Perform unsupervised learning using K-means clustering, Expectation-Maximization (EM) methods, and kernel tricks to recognize underlying patterns in experimental data as well as perform supervised learning using deep neural network modeling and backpropagation learning schemes Audience: Graduate
Sources
- Course catalogue ↗2026-2027