Medical Imaging MONAI Bootcamp

MONAI is a freely available, community-supported, open source PyTorch-based framework for deep learning in medical imaging. It provides domain-optimized foundational capabilities for developing medical imaging training workflows in a native PyTorch paradigm.

In collaboration with MICCAI Educational Initiative, we are excited to bring to the medical imaging researchers a pre-MICCAI 2020 event from September 30 to October 2 - the first ever ‘MONAI Bootcamp’ with training modules, an architectural deep dive and an open challenge. It is free to participate in this event and participants are provided with access to GPUs.

Prerequisites

Experience in applying deep learning to medical imaging (entry, intermediate, advanced level researchers are encouraged to join).
Experience with PyTorch Programming is not a requirement but beneficial to participants.

NVIDIA sponsors a GPU for the challenge winner.

MONAI bootcamp endorses the MICCAI SOCIETY CODE OF CONDUCT.

DETAILED AGENDA

Day 1: September 30, 2020 (7:30 AM - 11:30 AM PST)

Welcome & Introduction
Connect to the Hackathon cluster
Getting started with MONAI framework
- MONAI 101 kick start
  - MONAI technical overview presentation
  - Lab 1 - MONAI 101 workflow tutorial
- MONAI for PyTorch users
  - MONAI PyTorch paradigm presentation
  - Lab 2 - Bringing existing PyTorch implementations to MONAI
- Domain specialized transformation & persistent caching
  - How and why should researchers use domain optimized MONAI features
  - Lab 3 - Power of developing best practices using MONAI
- Distributed training with MONAI
  - Distributed training on multiple GPUs presentation
  - Lab 4 - Demonstration on multi-node setup and scalable performance
- Virtual social hour & mixer