Overview
This course covers modern techniques in medical image segmentation, focusing on deep learning and Automated Machine Learning (AutoML). The course aims to teach students how to apply deep convolutional neural networks for medical image segmentation, with a specific focus on 3D medical image segmentation. The course introduces a novel method that automatically considers and estimates various components of a deep neural network-based solution, leading to state-of-the-art performance on large-scale lesion segmentation datasets. The teaching method includes lectures on different topics in medical image segmentation, such as transformer-based networks, segmentation in federated learning, and shape priors in segmentation. This course is intended for individuals interested in the intersection of healthcare, deep learning, and medical imaging, particularly those looking to enhance their skills in medical image analysis and segmentation using advanced technologies.
Syllabus
Introduction
History of segmentation
Deep learning in segmentation
Neural Architecture Search
Multipath Search
Optimal Solutions
Recent Literature
Optimization
Beyond AutoML
Summary
Questions
Taught by
NVIDIA Developer