Pat Virtue

Assistant Teaching Professor
School of Computer Science , Carnegie Mellon University
Office: 6001 Gates Hillman Complex (GHC)

I'm an Assistant Teaching Professer in the Computer Science and Machine Learning departments here at CMU. As teaching faculty, I get to pour my heart and soul into courses, curriculum, and improving educational methods. This also means that am not running a research lab and am not planning to advise grad students in the near future. That being said, I'm always on the lookout for great TAs and anyone that wants to talk a little CS and AI education!

Current Teaching

Previous Teaching

Principles of Imperative Computation

CMU 15-122

Fall 2018 Instructor with co-instructor Iliano Cervesato, 500 students, and 39 TAs

Artifical Intelligence

UC Berkeley, CS 188
☆ Highest rating in course history: "How worthwhile was this course?"

Fall 2015 Instructor with co-instructor Stuart Russell, 400 students, and 9 TAs
Summer 2015   Instructor with 70 students and 3 TAs
Fall 2014 Teaching assistant with instructor Stuart Russell, 430 students, and 8 TAs

Machine Learning

UC Berkeley, CS 189
☆ Awarded EECS Outstanding Graduate Student Instructor Award

Spring 2015 Head TA with instructors Peter Bartlett and Alyosha Efros, 380 students, and 8 TAs

The Beauty & Joy of Computing

UC Berkeley, CS 10 (CS0 course)

6 terms,
Guest lecturer - Artificial Intelligence with instructors Dan Garcia, Justin Hsia, and Josh Hug

The Structure and Interpretation of Computer Programs

UC Berkeley, CS 61A (CS1 course)

Summer 2017 Tutor with instructors Stanfield Chu and Kevin Lin

GE Heathcare Advanced Course in Engineering

Edison Engineering Development Program, Advanced Course in Engineering

2004-2010 Instructor and curriculum designer
Image Reconstruction, Image Processing, and 2D/3D Medical Image Visualization


See full curriculum vitae

MRI Fingerprinting

Magnetic resonance fingerprinting is a revolutionary means to produce quantitative medical images from a single pseudorandom MRI scan. I am currently researching deep learning techniques to make MR fingerprinting more accurate, efficient, and versatile.


Complex-valued Neural Networks

Better than Real: Complex-valued Neural Nets for MRI Fingerprinting
Patrick Virtue, Stella X. Yu, Michael Lustig
International Conference on Image Processing 2017 paper / code


Predicting Image Quality

On the Empirical Effect of Gaussian Noise in Under-sampled MRI Reconstruction
Patrick Virtue, Michael Lustig
Tomography, in press paper



ESPIRiT—an Eigenvalue Approach to Autocalibrating Parallel MRI: Where SENSE Meets GRAPPA
Martin Uecker, Peng Lai, Mark J. Murphy, Patrick Virtue, Michael Elad, John M. Pauly, Shreyas S. Vasanawala, Michael Lustig
Magnetic Resonance in Medicine 2014 paper / code


MRI Cartilage Thickness

Validation of Cartilage Thickness Calculations Using Indentation Analysis
Matthew F. Koff, Le Roy Chong, Patrick Virtue, Dan Chen, Xioanan Wang, Timothy Wright, Hollis G. Potter
Journal of Biomechanical Engineering 2010 paper


MRI Growth Plate Analysis

Correlation of Magnetic Resonance Imaging and Histologic Examination of Physeal Bars in a Rabbit Model
Matthew F. Koff, Le Roy Chong, Patrick Virtue, Liang Ying, Purushottam A. Gholve, Scott A. Rodeo, Roger F. Widmann, Hollis G. Potter
Journal of Pediatric Orthopedics 2010 paper

Raptors Love Turkey Legs