Skip to main content

laying_tracks.gif

This is a brand new course and the machine learning field is moving incredibly fast. Please be patient with me while I work to maximize the utility of this course, and work out the new course bugs. The course will be updated as the semester progresses, as new material becomes available and based on student feedback. Please do not hesitate to contact me with any questions or concerns.
Welcome to the our brand new course on data science and machine learning for the medical physicist The course will be taught by Dr. Rafe McBeth, a medical physicist at the University of Pennsylvania,you will learn about the following topics.:
  • Interacting with common data found in the Medical Physics field (Radiology and Radiation Oncology)
  • Processing this data for use in machine learning models
  • Creating basic machine learning models for use in Radiology and Radiation Oncology
The overall goal of this course is to provide practical skills, such that the students can contribute to future clinical issues that benefit from the use of data science and machine learning. The majority of the didactic (lecture) material will be provided as uploaded videos that can be watched at the students convenience. The dedicated course time will be used as live lab coding sessions, where the students will work on the course material in a live coding environment. Since the majority of the learning material will be presented online, and the class size is small, students are expected to attend class/lab sessions in person and work together to solve the coding problems. I believe the in person lab sessions will be greatly beneficial to the students, and will help to foster a collaborative learning environment and increase the help I can provide for any computer and coding issues. Students that cannot attend a specific day can the instructor to discuss alternative arrangements. Logistic details for the course are as follows:
  • Course: MPHY 6120-01 Data Science & Artificial Intelligence for the Medical Physicist
  • Offered: Thursdays, 1:45 - 4:45 PM
  • Classroom: SCTR 10-146AB
  • Expected enrollment: 10

If you are a newcomer to Python or if you need a refresher, we recommend that you enroll in Penn’s free Introduction to Python Programming course on Coursera. Another useful resource is this recitation on Debugging and Profiling in Python by Arun Kirubarajan.

Quiz 1 "Course start, policies and basics" has been released. It is due before 11:59PM on Thursday, January 12, 2023.
Course number
MPHY 6120 - Data Science and Artificial Intelligence for Medical Physics
Prerequeisites
Students should feel confident creating python programs and using git.
Instructor
Rafe McBeth, PhD
Office
4-346W
Discussion Forum
XXX
Time and place
In Spring 2022, the in person section of the course will be in person
All lecture materials will be recorded and uploaded. Students are expected to watch this on their own time before meeting in person.
The first day of class is Thursday, January 12, 2023.
The first day of office hours will be January 15, 2023.
The final day of class is Thursday, May 7, 2023.
  • The class meets on Thursdays from 0:45-4:45 pm Eastern.

Guest lectures will be presented by other members of the Penn Medical Physics team as needed to enhance content. (TBD)

Course Description: The fields of data science (DS) and artificial intelligence (AI) are growing rapidly and will impact the future practice of medicine and the field of Medical Physics. This course is designed to enhance the graduate student’s ability to contribute to Medical Physics research and the integration of DS and AI into clinical practice. The course will teach students practical programming techniques (Python), data science infrastructure, data analysis workflows, medical physics specific programming and how to build, analyze and use machine learning (ML) and Deep learning (DL) AI models.

Course Format: Short on demand lectures will be provided for content being covered that week. It is expected that students will watch this material on their own. Meeting times will be dedicated to clarifying the lecture material, answering questions and working through hands on examples.

Required course resources: No textbook required. Course materials will be provided online and will include videos, slides, Jupyter notebooks and a GitHub repository for skeleton code and examples.

Course GitHub repository: https://github.com/MedPhysDS-AI Course website: https://medphysds-ai.github.io/

Optional course resources:

  • Machine and Deep Learning in Oncology, Medical Physics and Radiology by Issam El Naqa et al, second edition, 2022 ISBN 978-3-030-83046-5
  • How to Think Like a Computer Scientist: Learning with Python 2ed by Jeffrey Elkner, Allen B. Downey and Chris Meyers (Open Book Project) [https://greenteapress.com/wp/think-python-2e/]
  • Python for Data Analysis: Data Wrangling with Pandas, NumPy, and Jupyter 3rd Edition by Wes McKinney (Author) [ISBN-10: 109810403X]
Group Office Hours
TBD (Rafe McBeth)
Textbooks
Required: None currently required.
Grading
  • 50% Homework Assignments and quizzes
  • 25% for midterm project
  • 25% for final project
Score Grade
≥ 97 A+
93-97 A
90-93 A-
87-90 B+
83-87 B
80-83 B-
75-80 C+
70-75 C
65-70 C-
50-65 D
below 50 F