This picture showsAnnika Liebgott

M. Sc.

Annika Liebgott

Research Assistant
Institute for Signal Processing and System Theory

Contact

0049 711 685-67359
0049 711 685-67311

Pfaffenwaldring 47
70550 Stuttgart
Germany
Room: 2.275

Office Hours

by arrangement

Subject

My research is performed in cooperation with the Department of Radiology at the University Hospital of Tübingen and focuses on medical image processing involving machine learning. One key aspect is the comparisson of the performance of different machine learning techniques on various medical tasks. Methods involve e.g.:

  • Feature extraction and analysis
  • Feature-based classification, e.g. support vector machine or random forest
  • Deep learning methods, espeically CNN architectures
  • Active learning

Medical applications mainly belong to the area of oncology. Besides smaller tasks, there are two main research projects:

Automated segmentation of lung tumors in PET/CT images

Various machine learning techniques are employed to segment the primary lung tumors of patients suffering from non-small cell lung cancer. The data used in this project consists of images from positron emission tomography (PET) and computed tomography (CT).

The research focuses mainly on segmentation using as few training images as possible, i.e. techniques like active learning are investigated.

Prediction of therapy response of advanced melanoma patients receiving immunotherapy from PET/MR images using machine learning

Malignant melanoma is is caracterized by fast and wide spread in its advanced stages. Conventional therapies like chemo or radiation therapy are often hardly effective, leading to high recurrence and low 10 year survival rates. One major breakthrough in the therapy of melanoma could recently be accomplished by employing immune checkpoint inhibitor therapy, which is intended to stimulate the patients' immune system to attack cancerous cells. Since the first immunotherapy drugs have been cleared for clinical use in melanoma treatment in 2011, overall outcome in advanced-stage melanoma patients has increased significantly. Unfortunately, not all patients respond to the immune checkpoint inhibitors and it is not yet possible to determine whether immunotherapy is effective before starting the treatment.

The goal of this project is the development of a non-invasive Method to predict individual therapy response of patients suffering from metastatic melanoma using machine learning and routinely acquired radiological images. The data used in this work are combined positron emission tomography (PET) and magnetic ressonance imaging (MRI) images of patients with known therapy response, which were acquired at various time points over the course of treatment to monitor treatment response.

This work is supported by Vector Stiftung

Logo_Vector_Stiftung_RGB_300px

Student Theses (BA/FA/SA/MA)

It is generally possible to do a student thesis on both research projects, as well as on some smaller side projects that are avaiblable from time to time. The individual topics for a thesisand which types of machine learning techniques are employed depend on the questions currently arising in the projects and the type of student thesis. Depending on the task at hand, programming skills in Matlab and/or Python are neccessary, for theses involving deep learning techniques knowledge in tensorflow is beneficial. Whether or not a thesis topic can be offered strongly depends on the current number of active student theses and the existence of a task suitable for the desired thesis type.

ImFEATbox: a Matlab toolbox for image feature extraction and analysis

The toolbox, which is available on GitHub, includes varios algorithms for feature extraction and reduction. It has been developed at ISS and the University Hospital of Tübingen and has been employed for several (medical) image processing tasks: ImFEATbox

 

2020

Annika Liebgott, Tobias Haueise, Konstantin Nikolaou, Bin Yang and Sergios Gatidis
A Radiomics-based Approach to Predict Therapy Response in Advanced Melanoma Patients treated with Immunotherapy
Proceedings of the 37th Annual Scientific Meeting ESMRMB 2020, Barcelona, Spain, September 2020

Tobias Haueise, Annika Liebgott, Tobias Hepp, Konstantin Nikolaou, Bin Yang and Sergios Gatidis
Can Deep Learning-Based Feature Selection Improve Results in Radiomics? A Study on the Example of Gleason Score Prediction
Proceedings of the 37th Annual Scientific Meeting ESMRMB 2020, Barcelona, Spain, September 2020 

Annika Liebgott, Sergios Gatidis , Viet Chau Vu , Tobias Haueise, Konstantin Nikolaou and Bin Yang
Feature-based Response Prediction to Immunotherapy of late-stage Melanoma Patients Using PET/MR Imaging
Proceedings of the IEEE European Signal Processing Conference EUSIPCO 2020, Amsterdam, Netherlands, August 2020

T. Hepp, A. Othman, A. Liebgott, J. H. Kim, C. Pfannenberg and S. Gatidis
Effects of simulated dose variation on contrast-enhanced CT-based radiomic analysis for Non-Small Cell Lung Cancer
European Journal of Radiology, January 2020, doi: 10.1016/j.ejrad.2019.108804

 

2019

A. Liebgott, T. Haueise, T. Küstner, K. Nikolaou, B. Yang, S. Gatidis
Active Learning for the Adaption of Trained CNN Models for detection of Motion Artifacts to new Data
Proceedings of the 36th Annual Scientific Meeting ESMRMB 2019, Rotterdam, Netherlands, October 2019

A. Liebgott, D. Hinderer, K. Armanious, A. Bartler, K. Nikolaou, S. Gatidis and B. Yang
Prediction of FDG uptake in Lung Tumors from CT Images Using Generative Adversarial Networks
Proceedings of the IEEE European Signal Processing Conference EUSIPCO 2019, A Coruna, Spain, September 2019,
 
A. Liebgott, J. Yi, T. Küstner, K. Nikolaou, B. Yang, and S. Gatidis
Reinforcement Learning for Automated Reference-free MR Image Quality Assessment.
Proceedings of the Annual Meeting of the International Society for Magnetic Resonance in Medicine (ISMRM) 2019, Montreal, Canada, May 2019.

A. Liebgott, J. Steyer-Ege, T. Hepp, T. Küstner, K. Nikolaou, B. Yang and S. Gatidis
Feature Reduction and Selection: a Study on their Importance in the Context of Radiomics.
Proceedings of the Annual Meeting og the International Society for Magnetic Resonance in Medicine (ISMRM) 2019, Montreal, Canada, May 2019.
 

2018

A. Liebgott, T. Küstner, H. Strohmeier, T. Hepp, P. Mangold, P. Martirosian, F. Bamberg, K. Nikolaou, B. Yang and S. Gatidis
ImFEATbox: a toolbox for extraction and analysis of medical image features
International Journal of Computer Assisted Radiology and Surgery, (2018), Volume 13, Issue 12, pp 1881–1893. Springer Verlag, ISSN 861-6410, doi: https://link.springer.com/article/10.1007%2Fs11548-018-1859-7

Küstner, T., Gatidis, S., Liebgott, A., Schwartz, M., Mauch, L., Martirosian, P., Schmidt, H., Schwenzer, N., Nikolaou, K., Bamberg, F., Yang, B. & Schick, F.
A machine-learning framework for automatic reference-free quality assessment in MRI.
Magnetic Resonance Imaging, 53, 134 - 147. doi: https://www.sciencedirect.com/science/article/pii/S0730725X18302893?via%3Dihub 2018.
 
Milde, S., Liebgott, A., Wu, Z., Feng, W., Yang, J., Mauch, L., Martirosian, P., Bamberg, F., Nikolaou, K., Gatidis, S., Schick, F., Yang, B. & Küstner, T.
Graphical User Interface for Medical Deep Learning - Application to Magnetic Resonance Motion Artifact Detection
Proceedings of the IEEE Asia-Pacific Signal and Information Processing Association (APSIPA), Honolulu, Hawaii, 2018.
 

A. Liebgott, D. Boborzi, S. Gatidis, F. Schick, K. Nikolaou, B. Yang and T. Küstner
Active learning for automated reference-free MR image quality assessment: decreasing the number of required training samples by reduction of intra-batch redundancy
Proceedings of the joint Annual Meeting ISMRM-ESMRMB 2018, June 2018, Paris, France.

Küstner, T., Jandt, M., Liebgott, A., Mauch, L., Martirosian, P., Bamberg, F., Nikolaou, K., Gatidis, S., Yang, B. & Schick, F.
Motion artifact quantification and localization for whole-body MRI
Proceedings of the joint Annual Meeting ISMRM-ESMRMB 2018, June 2018, Paris, France.

Küstner, T., Liu, K., Liebgott, A., Mauch, L., Martirosian, P., Bamberg, F., Nikolaou, K., Yang, B., Schick, F. & Gatidis, S.
Simultaneous detection and identification of MR artifact types in whole-body imaging
Proceedings of the joint Annual Meeting ISMRM-ESMRMB 2018, June 2018, Paris, France.

A. Liebgott, S. Gatidis, F. Liebgott, K. Nikoalou and B. Yang
Automated Detection of High FDG Uptake Regions in CT Images
Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2018, April 2018, Calgary, Canada.

Küstner, T., Jandt, M., Liebgott, A., Mauch, L., Martirosian, P., Bamberg, F., Nikolaou, K., Gatidis, S., Schick, F. and Yang, B.
Automatic Motion Artifact Detection for Whole-Body Magnetic Resonance Imaging
Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2018, April 2018, Calgary, Canada.

Liebgott, A., Milde, S., Jandt, M., Mauch, L., Martirosian, P., Bamberg, F., Schick, F., Nikolaou, K., Yang, B., Gatidis, S. and Küstner, T.
Impact of Labeling Process on Automated Motion Artifact Detection in Whole-Body MR Images with a Deep Learning Approach: A Comparative Study
Proceedings of the ISMRM Workshop on Machine Learning, March 2018, Pacific Grove, CA, USA.

2017

T. Küstner, A. Liebgott, L. Mauch, P. Martirosian, F. Bamberg, K. Nikolaou, B. Yang, F. Schick, S. Gatidis
Automated reference-free detection of motion artifacts in magnetic resonance images
Magnetic Resonance Materials in Physics, Biology and Medicine, 2017.

T. Küstner, A. Liebgott, L. Mauch, P. Martirosian, F. Schick, F. Bamberg, K. Nikolaou, B. Yang, S. Gatidis
Automatic reference-free motion artifact detection and quantification in T1-weighted MR images of the head and abdomen
Proceedings of the Annual Scientific Meeting ESMRMB, October 2017, Barcelona, Spain.

A. Liebgott, S. Gatidis, P. Martirosian, F. Schick, B. Yang and T. Küstner
ImFEATbox: An MR Image Processing Toolbox for Extracting and Analyzing Features
Proceedings of the Annual Meeting ISMRM 2017, April 2017, Honolulu, Hawaii, USA.

T. Küstner, A. Liebgott, L. Mauch, P. Martirosian, K. Nikolaou, F. Schick, B. Yang and S. Gatidis
Automatic reference-free detection and quantification of MR image artifacts in human examinations due to motion
Proceedings of the Annual Meeting ISMRM 2017, April 2017, Honolulu, Hawaii, USA.

S. Gatidis, A. Liebgott, M. Schwartz, P. Martirosian, F. Schick, K. Nikolaou, B. Yang and T. Küstner
Automated reference-free assessment of MR image quality using an active learning approach: comparison of Support Vector Machine versus Deep Neural Network classification
Proceedings of the Annual Meeting ISMRM 2017, April 2017, Honolulu, Hawaii, USA.

T. Küstner, P. Wolf, M. Schwartz, A. Liebgott, F. Schick, S. Gatidis and B. Yang
An easy-to-use image labeling platform for automatic Magnetic Resonance Image Quality Assessment
Proceedings of the 14th IEEE International Symposium on Biomedical Imaging ISBI 2017, April 2017, Melbourne, Australia.

2016

S. Gatidis, A. Liebgott, H. Schmidt, NF. Schwenzer, P. Martirosian, K. Nikolaou, F. Schick, B. Yang and T. Küstner
Automated image quality assessment in whole-body MRI
Proceedings of the Annual Meeting RSNA, November 2016, Chicago, USA.

T. Küstner, M. Schwartz, A.Liebgott, P. Martirosian, S. Gatidis, NF. Schwenzer, F. Schick, H. Schmidt and B. Yang
An Active Learning platform for automatic MR image quality assessment
Proceedings of the Annual Meeting ISMRM 2016, May 2016, Singapore.

A. Liebgott, T. Küstner, S. Gatidis, F. Schick and B. Yang
Active Learning for Magnetic Resonance Image Quality Assessment
Proceedings of the 41th IEEE International Conference on Acoustics, Speech and Signal Processing ICASSP 2016, March 2016, Shanghai, China.

To the top of the page