Dr Steffen Bollmann

Senior Research Fellow

School of Electrical Engineering and Computer Science
Faculty of Engineering, Architecture and Information Technology
s.bollmann@uq.edu.au
+61 7 334 60360

Overview

Dr Steffen Bollmann joined UQ’s School of Electrical Imaging and Computer Science in 2020 where he leads the Computational Imaging Group. The Group is developing computational methods to extract clinical and biological insights from magnetic resonance imaging (MRI) data. The aim is to make cutting-edge algorithms and tools available to a wide range of clinicians and researchers. This will enable better images, faster reconstruction times and the efficient extraction of clinical information to ensure a better understanding of a range of diseases. Dr Bollmann was appointed Artificial Intelligence (AI) lead for imaging at UQ’s Queensland Digital Health Centre (QDHeC) in 2023.

His research expertise is in quantitative susceptibility mapping, image segmentation and software applications to help researchers and clinicians access data and algorithms.

Dr Bollmann completed his PhD on multimodal imaging at the University Children’s Hospital and Swiss Federal Institute of Technology (ETH) Zurich, Switzerland.

In 2014 he joined the Centre for Advanced Imaging at UQ as a National Imaging Facility Fellow, where he pioneered the application of deep learning methods for quantitative imaging techniques, in particular Quantitative Susceptibility Mapping.

In 2019 he joined the Siemens Healthineers collaborations team at the MGH Martinos Center in Boston on a one-year industry exchange where he worked on the translation of fast imaging techniques into clinical applications.

Research Interests

  • Quantitative Susceptibility Mapping
    Developing new methods to increase the robustness of processing quantitative susceptibility mapping.
  • Reproducible Research Software
    Developing software to enable reproducible neuroimaging, such as www.Neurodesk.org
  • Computational Imaging
    Developing tools to make computational algorithms for medical imaging more accessible and robust.
  • Image Segmentation
    Developing new methods to segment medical imaging data to extract quantitative information.

Research Impacts

Strong industry collaborations to bring research algorithms into applications such as Quantitative Susceptibility Mapping with industry partner Siemens Healthineers and the Neurodesk project with industry partner Oracle Cloud.

Further information is available at www.mri.sbollmann.net and regular research updates can be found on linkedin (https://www.linkedin.com/in/steffen-bollmann-00725097/) mastodon (https://masto.ai/@Sbollmann_MRI) and twitter/X (https://twitter.com/sbollmann_mri)

Qualifications

  • PhD, Swiss Federal Institute of Technology, Zurich

Publications

  • Barden, Anne, Shinde, Sujata, Beilin, Lawrence J., Phillips, Michael, Adams, Leon, Bollmann, Steffen and Mori, Trevor A. (2024). Adiposity associates with lower plasma resolvin E1 (Rve1): a population study. International Journal of Obesity. doi: 10.1038/s41366-024-01482-x

  • Domínguez D, Juan F, Stewart, Ashley, Burmester, Alex, Akhlaghi, Hamed, O'Brien, Kieran, Bollmann, Steffen and Caeyenberghs, Karen (2024). Improving quantitative susceptibility mapping for the identification of traumatic brain injury neurodegeneration at the individual level. Zeitschrift für Medizinische Physik. doi: 10.1016/j.zemedi.2024.01.001

  • Renton, Angela I., Dao, Thuy T., Johnstone, Tom, Civier, Oren, Sullivan, Ryan P., White, David J., Lyons, Paris, Slade, Benjamin M., Abbott, David F., Amos, Toluwani J., Bollmann, Saskia, Botting, Andy, Campbell, Megan E. J., Chang, Jeryn, Close, Thomas G., Dörig, Monika, Eckstein, Korbinian, Egan, Gary F., Evas, Stefanie, Flandin, Guillaume, Garner, Kelly G., Garrido, Marta I., Ghosh, Satrajit S., Grignard, Martin, Halchenko, Yaroslav O., Hannan, Anthony J., Heinsfeld, Anibal S., Huber, Laurentius, Hughes, Matthew E. ... Bollmann, Steffen (2024). Neurodesk: an accessible, flexible and portable data analysis environment for reproducible neuroimaging. Nature Methods. doi: 10.1038/s41592-023-02145-x

View all Publications

Supervision

View all Supervision

Available Projects

  • This project provides a reproducible neuroimaging data processing platform based on software containers (docker and singularity). The student will be able to learn about container technology and add new features to the platform, like the support of GPUs for deep learning applications, the support for M1/Arm processors by using muli-architecture builds and by developing graphical user interfaces.

  • Convolutional neural networks are particulary well suited to solve a variety of inverse problems in medical imaging. This project is a great chance to get involved in the field of medical image processing using deep learning techniques from image reconstruction, registration to segmentation. Prior knowledge in Python, Tensorflow/Keras, Pytorch, and Linux shell scripting are recommended.

View all Available Projects

Publications

Journal Article

Conference Publication

Edited Outputs

Other Outputs

  • Stewart, Ashley Wilton, Goodwin, Jonathan, Richardson, Matthew, Robinson, Simon Daniel, O’Brien, Kieran, Jin, Jin, Barth, Markus and Bollmann, Steffen (2023). Deep-learning-enabled differentiation between intraprostatic gold fiducial markers and calcification in quantitative susceptibility mapping.

  • Bollmann, Steffen, Janke, Andrew, Marstaller, Lars, Reutens, David, O'Brien, Kieran and Barth, Markus (2017). GRE and QSM average 7T model. The University of Queensland. (Dataset) doi: 10.14264/uql.2017.178

  • Bollmann, Steffen, Janke, Andrew, Marstaller, Lars, Reutens, David, O'Brien, Kieran and Barth, Markus (2017). MP2RAGE T1-weighted average 7T model. The University of Queensland. (Dataset) doi: 10.14264/uql.2017.266

  • Bollmann, Steffen, Janke, Andrew, Marstaller, Lars, Reutens, David, O'Brien, Kieran and Barth, Markus (2017). Turbo Spin Echo average 7T model. The University of Queensland. (Dataset) doi: 10.14264/uql.2017.267

PhD and MPhil Supervision

Current Supervision

  • Doctor Philosophy — Principal Advisor

    Other advisors:

  • Doctor Philosophy — Principal Advisor

    Other advisors:

  • Doctor Philosophy — Associate Advisor

  • Doctor Philosophy — Associate Advisor

    Other advisors:

  • Doctor Philosophy — Associate Advisor

    Other advisors:

Completed Supervision

Possible Research Projects

Note for students: The possible research projects listed on this page may not be comprehensive or up to date. Always feel free to contact the staff for more information, and also with your own research ideas.

  • This project provides a reproducible neuroimaging data processing platform based on software containers (docker and singularity). The student will be able to learn about container technology and add new features to the platform, like the support of GPUs for deep learning applications, the support for M1/Arm processors by using muli-architecture builds and by developing graphical user interfaces.

  • Convolutional neural networks are particulary well suited to solve a variety of inverse problems in medical imaging. This project is a great chance to get involved in the field of medical image processing using deep learning techniques from image reconstruction, registration to segmentation. Prior knowledge in Python, Tensorflow/Keras, Pytorch, and Linux shell scripting are recommended.