3D Medical Image Analysis with PyTorch

Image translation, CNN, Regression task

Jacob Reinhold

5 weeks · 8-10 hours per week

In this liveProject, you’ll take on the role of a machine learning engineer at a healthcare imaging company, processing and analyzing magnetic resonance (MR) brain images. Your current medical image analysis pipelines are set up to use two types of MR images, but a new set of customer data has only one of those types! Your challenge is to build a convolutional neural network that can perform an image translation to provide you with your missing data. You’ll do this using the deep learning framework PyTorch and a large preprocessed set of MR brain images. The company also wants to make sure your image translation convolutional neural network reliably produces the desired MR image, so you will need to provide qualitative and quantitative results demonstrating your method’s effectiveness.

project author

Jacob Reinhold

Jacob Reinhold is a PhD student in electrical engineering at Johns Hopkins University. His research focuses on medical image analysis, specifically in applying deep learning techniques and theory to study anomaly detection in magnetic resonance and computed tomography images. His work has been published in peer-reviewed journals and conferences in the field. Prior to his graduate studies, he worked as an engineering scientist associate in the Applied Research Laboratories at the University of Texas at Austin and received his BS in electrical engineering at the University of Texas at Austin.

Prerequisites

This liveProject is for experienced Python programmers, familiar with object-oriented programming techniques and Python scientific computing packages. You will need to know the basics of machine learning and statistics, but this course will teach you the advanced techniques. Throughout, you’ll use the Google Collaboratory (“Colab”) coding environment to access free GPU computer resources and speed up your training times. To begin this liveProject, you will also need to be familiar with:

TOOLS

Basics of Matplotlib
Basics of Jupyter Notebook
Basics of Git
Intermediate PyTorch

TECHNIQUES

Basics of gradient descent and SGD
Basics of Loss functions
Basics of Back-propagation
Basics of neural networks
Basics of advanced functions for ANNs such as softmax, sigmoid, ReLu

you will learn

The skills you develop in this liveProject are useful for all practitioners and researchers of deep learning. Utilizing the powerful PyTorch deep learning framework, you’ll learn techniques for computer vision that are easily transferable outside of medical imaging, such as depth estimation in natural images for self-driving cars, removing rain from natural images, and working with 3D data. You will learn:

How to load and process imaging data for deep learning applications
How to build a convolutional neural network
How to train a neural network for a regression task
How to evaluate the predictions of your neural network
How to handle and visualize medical imaging data

features

Self-paced: You choose the schedule and decide how much time to invest as you build your project.
Project roadmap: Each project is divided into several achievable steps.
Peer support: Chat with other participants within the liveProject platform.
Compare with others: For each step, compare your deliverable to the solutions by the author and other participants.
Book and video resources: Excerpts from Manning books and videos are included, as well as references to other resources.