Contact

Rüdiger Göbl, M.Sc. Mail: Technische Universität München Fakultät für Informatik I16 Boltzmannstr. 3 85748 Garching b. München Room: MI 03.13.044 Phone: +49 (89) 289-19417 Klinikum Rechts der Isar Interdisziplinäres Forschungslabor Ismaninger Str. 22 81675 München Room: Building 501 - 01.3a-c Phone: +49 89 4140 6457

Research Interests and Projects

• Fields of Interest
  • Interventional Ultrasound
  • Ultrasound Reconstruction
  • Computer Vision
• Research Project: Realtime Brain Shift Tracking and Tool Detection from Tri-Dimensional Ultrasound Data
• Master's Thesis: Parametric and Statistical Models for Computational Sonography

Active Research Projects

SUPRA: Software Defined Ultrasound Processing for Real-Time Applications SUPRA is an open-source pipeline for fully software defined ultrasound processing for real-time applications. Covering everything from beamforming to output of B-Mode images, SUPRA can help to improve the reproducibility of results and does allow for a full customization of the image acquisition workflow. Including all processing stages of a common ultrasound pipeline, it can be executed in 2D and 3D on consumer GPUs in real-time. Even on hardware as small as the CUDA enabled Jetson TX2, SUPRA allows for 2D imaging in real-time. You can access the code on our github page: https://github.com/IFL-CAMP/supra Additional information can be found in our work on SUPRA Göbl, R. and Navab, N. and Hennersperger, C., SUPRA: Open Source Software Defined Ultrasound Processing for Real-Time Applications, eprint arXiv:1711.06127, Nov 2017, under review for IPCAI2018 The development of SUPRA was partly funded by the European Horizon 2020 Project EDEN2020.

EDEN2020: Enhanced Delivery Ecosystem for Neurosurgery EDEN2020 (Enhanced Delivery Ecosystem for Neurosurgery) aims to develop the gold standard for one-stop diagnosis and treatment of brain disease by delivering an integrated technology platform for minimally invasive neurosurgery. A team of first-class industrial partners (Renishaw plc. and XoGraph ltd.), leading clinical oncological neurosurgery team (Università di Milano, San Raffaele and Politecnico di Milano) lead by Prof. Lorenzo Bello and the involvement of leading experts in shape sensing (Universitair Medisch Centrum Groningen) under supervision of Prof. Dr. Sarthak Misra The project is coordinated by Dr. Rodriguez y Baena, Imperial College London. His team provides the core technology for the envisioned system, the bendable robotic needle. During the course of EDEN2020 this interdisciplinary team will work on the integration of 5 key concepts, namely (1) pre-operative MRI and diffusion-MRI imaging, (2) intra-operative ultrasounds, (3) robotic assisted catheter steering, (4) brain diffusion modelling, and (5) a robotics assisted neurosurgical robotic product (the Neuromate), into a pre-commercial prototype which meets the pressing demand for better and less invasive neurosurgery. Our chair will be focusing on the imaging components (i.e. (1) and (2)), targeting the realtime compensation of tissue movement and accurate localization of the flexible catheters at hand. We will further extend the findings of FP7 ACTIVE, in which we successfully combined pre-operative MRI with intra-operative US through deformable 3D-2D registration, making us most qualified for this role.

Computational Sonography 3D ultrasound imaging has high potential for various clinical applications, but often suffers from high operator-dependency and the directionality of the acquired data. State-of-the-art systems mostly perform compounding of the image data prior to further processing and visualization, resulting in 3D volumes of scalar intensities. This work presents computational sonography as a novel concept to represent 3D ultrasound as tensor instead of scalar fields, mapping a full and arbitrary 3D acquisition to the reconstructed data. The proposed representation compactly preserves significantly more information about the anatomy-specific and direction-depend acquisition, facilitating both targeted data processing and improved visualization. We show the potential of this paradigm on ultrasound phantom data as well as on clinically acquired data for acquisitions of the femoral, brachial and antebrachial bone. Further investigation will consider additional compact directional-dependent representations on the one hand and on the other hand modify Computational Sonography from working on B-Mode images to RF-envelope statistics, motivated by the statistical process of image formation. We will show the advantages of the proposed improvements on simulated ultrasound data, phantom and clinically acquired ultrasound data.

Publications

2019
	M. Paschali, W. Simson, A. Guha Roy, R. Göbl, F. Naeem, C. Wachinger, N. Navab Manifold Exploring Data Augmentation with Geometric Transformations for Increased Performance and Robustness 26th international conference on Information Processing in Medical Imaging (IPMI), Hong Kong 2019. The original publication is available currently online on arXiv and will become available at link.springer.com (bib)
	F. Langsch, S. Virga, J. Esteban, R. Göbl, N. Navab Robotic Ultrasound for Catheter Navigation in Endovascular Procedures IEEE/RSJ International Conference on Intelligent Robots and Systems 2019 (bib)
	W. Simson, R. Göbl, M. Paschali, M. Krönke, K. Scheidhauer, S. Weber, N. Navab End-to-End Learning-Based Ultrasound Reconstruction The first two authors contributed equally. The original manuscript is available currently online on arXiv. (bib)
	J. Rackerseder, M. Baust, R. Göbl, N. Navab, C. Hennersperger Landmark-Free Initialization of Multi-Modal Image Registration Bildverarbeitung fuer die Medizin (BVM), Luebeck, March 17-19, 2019. (bib)
	M. Esposito, C. Hennersperger, R. Göbl, L. Demaret, M. Storath, N. Navab, M. Baust, A. Weinmann Total Variation Regularization of Pose Signals with an Application to 3D Freehand Ultrasound IEEE Transactions on Medical Imaging (TMI), Volume 38, Issue 10, pp 2245-2258, October 2019. (bib)
2018
	J. Rackerseder, M. Baust, R. Göbl, N. Navab, C. Hennersperger Initialize globally before acting locally: Enabling Landmark-free 3D US to MRI Registration 21st International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI), Granada, Spain, September 2018 (bib)
	R. Göbl, N. Navab, C. Hennersperger SUPRA: open-source software-defined ultrasound processing for real-time applications International Journal of Computer Assisted Radiology and Surgery / 9th International Conference on Information Processing in Computer-Assisted Interventions (IPCAI), Berlin, Germany, June 2018. The original publication is available online at link.springer.com (bib)
	S. Virga, R. Göbl, M. Baust, N. Navab, C. Hennersperger Use the Force: Deformation Correction in Robotic 3D Ultrasound International Journal of Computer Assisted Radiology and Surgery / 9th International Conference on Information Processing in Computer-Assisted Interventions (IPCAI), Berlin, Germany, June 2018. The original publication is available online at springer.com (bib)
2017
	R. Göbl, S. Virga, J. Rackerseder, B. Frisch, N. Navab, C. Hennersperger Acoustic window planning for ultrasound acquisition International Journal of Computer Assisted Radiology and Surgery / 8th International Conference on Information Processing in Computer-Assisted Interventions (IPCAI), Barcelona, Spain, June 2017. The original publication is available online at link.springer.com (bib)

Teaching

• Lecture Basic Mathematical Tools for Imaging and Visualization (2018WiSe)
• Lecture Grundlagen: Algorithmen und Datenstrukturen (2018SoSe)
• Lecture Basic Mathematical Tools for Imaging and Visualization (2017WiSe)
• Lecture Algorithmen und Datenstrukturen (für ET/IT) (2017SoSe)
• Praktikum Project Management and Software Development for Medical Applications (2016WiSe)
• Hauptseminar Image-based tool tracking for computer aided biopsy procedures (2016WiSe)
• Praktikum Project Management and Software Development for Medical Applications (2016SoSe)

Finished Student Projects

• Theses
  • Catheter tracking in 3D ultrasound with fully convolutional neural networks (MSc)
  • 3D Freehand Ultrasound Calibration with Convolutional Networks (MSc)