Deep Learning for Medical Image Analysis

Paul F. Jaeger, Fabian Isensee, Jakob Wasserthal, Jens Petersen, David Zimmerer (Medical Image Analysis Lab at DKFZ)

March 20, 2018

Abstract

The field of medical image analysis (MIA) has changed rapidly in recent years due to the successes of deep learning. Despite being one of the most promising fields of applications for deep learning methods, many challenges arise when trying to adapt state-of-the-art models to the medical imaging domain including the handling of small data sets, unlabeled data, volumetric data, noisy data, noisy annotations, and so on.

This series of talks is organized by the MIA lab at the German Cancer Research Center and gives an overview of current deep learning methods and how they are applied in MIA.It is composed of five parts: Classification, Segmentation, Object Detection, Generative Models and Semi-Supervised Learning. Each part starts off with a thorough motivation, shows exemplary use cases related to MIA, provides a brief model overview and describes the current state-of-the-art methods in the respective area.

Part I: Classification, Segmentation & Object Detection, 20 March 2018

Intro

Image Classification, speaker: Jakob Wasserthal

Segmentation, speaker: Fabian Isensee

Object Detection, speaker: Paul F. Jaeger

Part II: Deep Generative Models & Semi-Supervised Learning, 27 March 2018

Deep Generative Models, speaker: Jens Petersen

Semi-Supervised Learning, speaker: David Zimmerer