Hauptseminar/Master Seminar: Deep Learning for Medical Applications

Prof. Nassir Navab, Shadi Albarqouni, Christoph Baur

Type: Advanced Seminar Course Module IN8901
Type: Master Seminar Module IN2107
SWS: 2+0
ECTS: 5 Credits
Location: CAMP Seminar Room, 03.13.010
Time: 12:30 - 14:00 (except for the introductory meeting!)
Course Language: English

Announcements

• 29-05-2017: You can submit/push your report here. Please request an access first. Deadline is 28th July 2017 at 23:59.
• 11-04-2017: Papers are assigned to the tutors. Please contact your tutor (latest 2 weeks before your presentation) to discuss with her/him the structure of your presentation.
• 03-04-2017: Please send us your preferences (at least three papers) by Monday, 10th April 2017.
• 01-02-2017: If you are still interested in the seminar, please write a brief motivation paragraph (few lines) showing your interest and your background in Machine/Deep Learning. Send this email, with a title “DLMA_Motivation”, to Shadi Albarqouni (shadi.albarqouni@tum.de) till 08.02.2017
• 20-01-2017: Registration is open from 03.02.2017 to 08.02.2017 through the TUM Matching Platform.
• 20-01-2017: Tentative Schedule is up!
• 18-01-2017: An introductory meeting: Wednesday, 1 February 2017 at 2 PM (30 minutes) in CAMP Seminar Room, 03.13.010.
• 09-01-2017: Website is up!

Introduction

• Deep Learning is growing tremendously in Computer Vision and Medical Imaging as well. Highly impacted journals in medical imaging community, i.e. IEEE Transaction on Medical Imaging, published recently their special edition on Deep Learning [1]. The Seminar will propose a list of recent scientific articles related to the main current research topics in deep learning for Medical Applications together with some interesting papers from other communities.

Registration

• It should be done through the TUM Matching Platform (please pay attention to the Deadlines).

Requirements

In this Master Seminar (formerly Hauptseminar), each student is asked to send three preferences from the list, then he will be assigned one paper. In order to successfully complete the seminar, participants have to fulfill these requirements:

• Presentation: The selected paper is presented to the other participants (20 minutes presentation 10 minutes questions). Use the CAMP templates for PowerPoint camp-tum-jhu-slides.zip, or Latex: CAMP-latex-template.zip.
• Written Report: A document of maximum 2 pages should be submitted before the deadline. A link to the shared repository will be announced later.
• Attendance: Participants have to participate actively in all seminar sessions.

The students are required to attend each seminar presentation which will be held during this course. Each presentation is followed by a discussion and everyone is encouraged to actively participate. The report must include all references used and must be written completely in your own words. Copy and paste will not be tolerated. Both report and presentation have to be done in English.

Schedule

Date	Time	Topic	Slides	Students
01.02.2017	14:00	Introductory meeting	Slides	Students who join this meeting and show their interest will have high priority!
27.04.2017	12:30 - 14:00	Kick-Off Session	Review	CNN (), RNN (Gerome/Ahmad), Autoencoders (Abhijit)
11.05.2017	12:30 - 14:00	Presentation Session 1		Ghazal, Ishmeet, Taha
18.05.2017	12:30 - 14:00	Presentation Session 2		Agnieszka, Jens, Yatin
01.06.2017	12:30 - 14:00	Presentation Session 3		Tomislav, Daniel, Anil
08.06.2017	12:30 - 14:00	Presentation Session 4		Evangelos, Edward, Fang
06.07.2017	12:30 - 14:00	Presentation Session 5		Cristian Ioan, Mustafa Taha
13.07.2017	12:30 - 14:00	Presentation Session 6		Can, Anne-Marie, Abu Rasiuddin

List of Topics and Material

The list of papers:

No	Application	Title	Conference/Journal	Tutor	Student	Link
1	CNN/FCN	Dermatologist-level classification of skin cancer with deep neural networks	Nature	Shadi	Ghazal	HTML
2	-	Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis	Nature	Marie	Ishmeet	HTML
3	-	CT Image Denoising with Perceptive Deep Neural Networks	arXiv	Christoph	Taha	PDF
4	-	CNN-based Segmentation of Medical Imaging Data	arXiv	Marie	Evangelos	PDF
5	-	The One Hundred Layers Tiramisu: Fully Convolutional DenseNets? for Semantic Segmentation	arXiv	Patrick	Can	PDF
6	-	Deep Image Retrieval: Learning global representations for image search	ECCV 2016			PDF
7	-	Deep Learning Computed Tomography	MICCAI 2016	Christoph	Anne-Marie	PDF
8	-	Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation	arXiv	Ahmad	Edward	PDF
9	-	DeepNAT?: Deep Convolutional Neural Network for Segmenting Neuroanatomy	arXiv	Ahmad	Fang	PDF
10	RNN/LSTM	Parallel Multi-Dimensional LSTM, With Application to Fast Biomedical Volumetric Image Segmentation	NIPS	Patrick	Abu Rasiuddin	PDF
11	-	Combining Fully Convolutional and Recurrent Neural Networks for 3D Biomedical Image Segmentation	NIPS	Patrick	Cristian Ioan	PDF
12	-	Prospective identification of hematopoietic lineage choice by deep learning	Nature	Tingying		Preview Link
13	-	Recurrent Models of Visual Attention	NIPS	Ahmad	Mustafa Taha	PDF
14	-	Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences	NIPS			PDF
15	GANs	Adversarial Networks for the Detection of Aggressive Prostate Cancer	arXiv	Sailesh	Tomislav	PDF
16	-	Medical Image Synthesis with Context-Aware Generative Adversarial Networks	arXiv			PDF
17	-	Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery	IPMI	Christoph	Daniel	PDF
18	-	Unsupervised domain adaptation in brain lesion segmentation with adversarial networks	IPMI			PDF
19	-	(Unsupervised*) Image-to-Image Translation with Conditional Adversarial Nets	arXiv	Shadi	Anil	PDF (PDF*)
20	SAE/CAE/VAE	Deep Patient: An Unsupervised Representation to Predict the Future of Patients from the Electronic Health Records	Nature	Shadi	Agnieszka	HTML
21	-	Medical image denoising using convolutional denoising autoencoders	arXiv	Marie	Yatin	PDF
22	-	Manifold Learning with Variational Auto-encoder for Medical Image Analysis	UNC	Christoph	Jens	PDF
23	RL	An artificial agent for anatomical landmark detection in medical images	MICCAI 2016			PDF
24	-	An Artificial Agent for Robust Image Registration	AAAI 2017			PDF

ICML: International Conference on Machine Learning
TMI-SIDL: IEEE Transaction on Medical Imaging, Special Issue on Deep Learning
ISBI: International Symposium on Biomedical Imaging
NIPS: Neural Information Processing Systems
ICRA: International Conference on Robotics and Automation
IJCARS: International Journal on Computer Assisted Radiology and Surgery
JBHI: IEEE Journal of Biomedical and Health Informatics
MICCAI: Medical Image Computing and Computer Assisted Intervention

Literature and Helpful Links

A lot of scientific publications can be found online.

The following list may help you to find some further information on your particular topic:

• Microsoft Academic Search
• Google Scholar
• CiteSeer
• CiteULike
• Collection of Computer Science Bibliographies

Some publishers:

• ScienceDirect (Elsevier Journals)
• IEEE Journals
• ACM Digital Library

Libraries (online and offline):

• http://rzblx1.uni-regensburg.de/ezeit/ (Elektronische Zeitschriften Bibliothek)
• Verbundkatalog des Bibliotheksverbundes Bayern (BVB)
• Computer ORG
• http://www.ub.tum.de/ (TUM Library)
  • To get access onto the electronic library, see http://www.ub.tum.de/medien/ejournals/readme.html
  • "proxy.biblio.tu-muenchen.de" mit Port 8080 (nur fuer http). Damit klappen zumindest portal.acm.org und computer.org meistens
• Various proceedings of conferences in our AR-Lab, 03.13.036 (These proceedings are not for lending!)

Some further hints for working with references:

• JabRef is a Java program for comfortable working with Bibtex literature databases. Handy feature: if you know the PubMed ID for an article, JabRef can import data from there (via "Web Search/Medline").
• Mendeley is a cross-platform program for organising your references.

If you find useful resources that are not already listed here, please tell us, so we can add them for others. Thanks.