Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

YouTube

On Bayesian Models with Networks for Reconstruction and Detection

Institute for Pure & Applied Mathematics (IPAM) via YouTube

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a 33-minute conference talk on Bayesian models with neural networks for medical image reconstruction and outlier detection. Delve into Ender Konukoglu's presentation from ETH Zurich at the Deep Learning and Medical Applications 2020 event, hosted by the Institute for Pure & Applied Mathematics at UCLA. Learn about network-based prior models for MRI reconstruction, the advantages of generative modeling, and unsupervised outlier detection techniques. Gain insights into topics such as image enhancement, posterior distribution, variational autoencoders, and Bayesian frameworks in medical image analysis. Examine experimental details and ROC curves to understand the practical applications of these advanced techniques in the field of medical imaging.

Syllabus

Intro
Outline
Acknowledgements
Examples: Image enhancement
Posterior distribution
Cartoon representation
Why use generative models for analyzing images?
Principal component analysis
Variational auto-encoders
MRI acquisition
Bayesian model for image reconstruction
MAP estimation with network prior
Advantage of generative modeling: decoupling
A distinction in the concept of "prior"
Unsupervised outlier detection
Restoration for outlier detection
Experimental details
ROC curves

Taught by

Institute for Pure & Applied Mathematics (IPAM)

Reviews

Start your review of On Bayesian Models with Networks for Reconstruction and Detection

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.