Modern Medical Image Segmentation, AutoML, and Beyond
Offered By: Nvidia via YouTube
Course Description
Overview
Explore cutting-edge techniques in medical image segmentation, AutoML, and advanced topics in this comprehensive talk by NVIDIA's Applied Research Scientist, Dong Yang. Delve into the evolution of image segmentation, the impact of deep learning, and the potential of Automated Machine Learning (AutoML) in enhancing model efficiency. Discover a novel method that systematically considers multiple components of deep neural network-based solutions for 3D medical image segmentation. Examine the proposed predictor-based AutoML algorithm and its large-scale neural architecture search space. Gain insights into state-of-the-art performance on lesion segmentation datasets and the method's transferability across different datasets. Explore additional topics such as transformer-based networks, federated learning, semi-supervised learning, and the integration of shape priors in segmentation. Conclude with a summary and Q&A session to deepen your understanding of modern medical image analysis techniques.
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
Tags
Related Courses
Advanced Deployment Scenarios with TensorFlowDeepLearning.AI via Coursera Big Data for Reliability and Security
Purdue University via edX Intel® Solutions Pro – AI in the Cloud
Intel via Coursera Responsible AI for Developers: Privacy & Safety
Google via Google Cloud Skills Boost Edge Analytics: IoT and Data Science
LinkedIn Learning