Image Guidance

Clinical navigation is largely based on image data and we focus on fast image modalities allowing for feedback and control. Particularly, we study optical coherence tomography (OCT) for precise, high resolution image guidance, e.g., in soft tissue and cardiac interventions. In addition we are interested in ultrasound (US) as a complementary modality for soft tissue imaging. We study methods to estimate tissue properties from OCT and US, e.g., to obtain forces and elasticity, and we integrate image guidance with robotics, navigation and machine learning.

Selected publications

N. Gessert, S. Latus, Y. S. Abdelwahed, D. M. Leistner, M. Lutz, A. Schlaefer (2019). Bioresorbable Scaffold Visualization in IVOCT Images Using CNNs and Weakly Supervised Localization. SPIE Medical Imaging 2019: Image Processing. Accepted.

N. Gessert, J. Beringhoff, C. Otte, A. Schlaefer (2018). Force Estimation from OCT Volumes using 3D CNNs. Int J Comput Assist Radiol Surg. 13 (7), 1073–1082.

S. Latus, F. Griese, M. Gräser, M. Möddel, M. Schlüter, C. Otte, N. Gessert, T. Saathoff, T. Knopp, A. Schlaefer (2018). Towards bimodal intravascular OCT MPI volumetric imaging. Proceedings of SPIE Medical Imaging: Physics of Medical Imaging 10573E.

R. Berndt, R. Rusch, L. Hummitzsch, M. Lutz, K. Heß, K. Huenges, B. Panholzer, C. Otte, A. Haneya, G. Lutter, A. Schlaefer, J. Cremer, J. Groß (2017). Development of a new catheter prototype for laser thrombolysis under guidance of optical coherence tomography (OCT): validation of feasibility and efficacy in a preclinical model. Journal of Thrombosis and Thrombolysis. 43 (3), 352-360.

O. Shahin, A. Beširevic, M. Kleemann, A. Schlaefer (2014). Ultrasound-based tumor movement compensation during navigated laparoscopic liver interventions. Surg Endosc. 28 (5), 1734-1741.

Motion Compensation

Motion, i.e., autonomic, active or pathological, is an important aspect that has to be considered when working with people. Predicting and compensating for autonomic motion is important for the quality of many applications like radiosurgery or transcranial magnetic stimulation. In addition, imaging quality can typically be improved if the motive's motion is known and compensated for. This is especially true for (bio-)medical imaging modalities like fluorescence microscopy and intravascular optical coherence tomography which picture very small structures.

We develop algorithms and systems to identify, predict, measure and finally compensate for motion. In addition, we study the reliability, robustness and treatment quality of existing and new methods.

Respiratory Motion Compensation for Robotic Radiosurgery

motion_comp_sketch.png

A tracking camera is used to measure the external motion which is correlated to the irregularly measured internal motion. Based on a prediction model the position of the linear accelerator (linac) is adapted to compensate for the respiratory motion.

Robotic radiosurgery is a technique, which typically uses high radiation doses for the treatment of cancer. To reduce side effects, the Opens internal link in new windowtreatment is planned in advance to be highly conformal, having steep dose gradients tightly following the tumor's contour. As respiratory motion introduces a significant displacement of organs like prostate, liver and lung, it has to be compensated during treatment. This is a two part problem:

  • Deriving the actual displacement of the tumor, e.g., from external motion using a correlation model.
  • Predicting the future movement of the tumor to account for inevitable delays in the system.

We study stopping criteria for respiratory motion compensation. In cooperation with the Institute of Software Systems, we developed a system to detect artifacts like coughing, sneezing or yawning alongside with other disturbances from regular breathing in respiratory data. The system is based on On-Line Model Checking (OMC), a dynamic technique that monitors a system, here a breathing model during runtime.

mot_comp_omc.png

Comparison of OMC with some state-of-the-art predictors for the means of stopping respiratory motion. Vertical lines denote the OMC evaluation results (green: valid, red: invalid) and red dots the error of the correlation model. While the very prominent artifact in a) is detected by OMC and (only) one predictor, the transition in b) is solely detected by OMC. Although the correlation model error goes up by about 1.5mm the prediction errors stay low.

Motion Compensation of the Head

MC_TMS.png

Experimental setup of TMS head motion compensation. (a) shows the depth image of the time of flight camera (e), (b) the head phantom mounted to a robot, (c) the coil mounted on a robot and (d) a tracking camera. The robot (b) is moving the head and the robot (c) moves the coil to compensate for this.

Transcranial magnetic stimulation (TMS) is an effective medical procedure for diagnostic and therapeutic purposes. Typically, the TMS procedure involves a physician who places the magnetic coil close to the patient's head to stimulate a specific cortex. Hence, it is important to compensate for any motion of the head to ensure correct stimulation and to avoid potentially harmful collisions between head and coil.

We develop a motion compensation system that compensates for head motion by tracking the head using (depth-)cameras and letting a robot move the coil accordingly. 

Motion Compensation for Intravascular Optical Coherence Tomography

IVOCT_CardiacSetup2.png

Setup for simulation of the arterial motion during IVOCT imaging. (a) Exemplary IVOCT image allocated within an human coronary artery. Seam line artifact (arrow) and guidewire artifact (asterisk) are highlighted. (b) Flexible vessel phantom integrated in the setup. (c) IVOCT catheter feeding. (d) Tracking camera. (e) Periodically triggered valve. (f) Water reservoir.

Intravascular Optical Coherence Tomography (IVOCT) supplements the diagnosis in percutaneous coronary interventions. Cross-sectional images of the arterial lumen are allocated and the luminal structure and dimension is assessed.

We implemented a setup to simulate the arterial motion during the cardiac cycle in order to analyze the motion artifacts in IVOCT images.

Cell motion compensation for fluorescence microscopy

When studying intra-cellular structures over longer time frames, i.e., propagation of calcium signals, knowing the motion as well as the deformation of the cell is of great importance for a detailed signal analysis.

We develop methods to analyze the motion and deformation of cells in order to compensate for this during analysis.

Contact

Opens internal link in new windowSarah Latus (sarah.latus(at)tuhh.de)

Opens internal link in new windowAlexander Schlaefer (schlaefer(at)tuhh.de)

References

  • O. Rajput, S.-T. Antoni, C. Otte, T. Saathoff, L. Matthäus, A. Schlaefer (2016). High Accuracy 3D Data Acquisition Using Co-Registered OCT and Kinect. 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems Baden-Baden [www] [BibTex]

  • S.-T. Antoni, J. Rinast, X. Ma, S. Schupp, A. Schlaefer (2016). Online model checking for monitoring surrogate based respiratory motion tracking in radiation therapy. CARS 2016 Proceedings, supplement of the International Journal of CARS accepted. [Abstract] [BibTex]

  • S.-T. Antoni, X. Ma, S. Schupp, A. Schlaefer (2016). Reducing false discovery rates for on-line model-checking based detection of respiratory motion artifacts. Gemeinsamer Tagungsband der Workshops der Tagung Software Engineering 2016 (SE 2016), Wien, Feb. 2016. [Abstract] [BibTex]

  • A. Tack, Y. Kobayashi, T. Gauer, A. Schlaefer, R. Werner (2015). Groupwise Registration for Robust Motion Field Estimation in Artifact-Affected 4D CT Images.. Workshop on Imaging and Computer Assistance in Radiation Therapy, MICCAI 2015 [BibTex]

  • D. Schetelig, I. M.A. Wolf , B.-P. Diercks, R. Fliegert, A. H. Guse, A. Schlaefer, R. Werner (2015). A Modular Framework for Post-Processing and Analysis of Fluorescence Microscopy Image Sequences of Subcellular Calcium Dynamics. Bildverarbeitung für die Medizin 2015, Algorithmen - Systeme - Anwendungen. Proceedings des Workshops vom 15. bis 17. März 2015 in Lübeck Springer: 401-406. [Abstract] [doi] [BibTex]

  • S.-T. Antoni, A. Dabrowski, D. Schetelig, B.-P. Diercks, R. Fliegert, R. Werner, I. Wolf, A. H. Guse, A. Schlaefer (2015). Segmentation of T-cells in fluorescence microscopy. In Proc. IEEE Engineering in Medicine and Biology Society (EMBC'15) Milan, Italy [Abstract] [www] [BibTex]

  • S.-T. Antoni, J. Rinast, S. Schupp, A. Schlaefer (2015). Comparing Model-free Motion Prediction and On-line Model Checking for Respiratory Motion Management. Gemeinsamer Tagungsband der Workshops der Tagung Software Engineering Dresden, Germany 15-18. [Abstract] [www] [BibTex]

  • S.-T. Antoni, J. Rinast, S. Schupp, A. Schlaefer (2015). Evaluation des Einflusses von Artefakten auf den Korrelationsfehler in der bewegungskompensierten Radiochirurgie. Tagungsband der 14. Jahrestagung der Deutschen Gesellschaft für Computer- und Roboterassistierte Chirurgie Bremen, Germany 133-138. [Abstract] [BibTex]

  • S.-T. Antoni, R. Plagge, R. Dürichen, A. Schlaefer (2015). Detecting Respiratory Artifacts from Video Data. In Handels, Heinz and Deserno, Thomas Martin and Meinzer, Hans-Peter and Tolxdorff, Thomas (Eds.) Bildverarbeitung für die Medizin 2015 Algorithmen - Systeme - Anwendungen. Proceedings des Workshops vom 15. bis 17. März 2015 in Lübeck Springer Berlin Heidelberg: 227-232. [Abstract] [doi] [www] [BibTex]

  • N. Lessmann, D. Drömann, A. Schlaefer, (2014). Feasibility of respiratory motion-compensated stereoscopic X-ray tracking for bronchoscopy. Int J Comput Assist Radiol Surg. 9 (2), 199-209. [doi] [BibTex] [pmid]

  • O. Shahin, A. Beširevi?, M. Kleemann, A. Schlaefer (2014). Ultrasound-based tumor movement compensation during navigated laparoscopic liver interventions. Surg Endosc. 28 (5), 1734-1741. [doi] [BibTex]

  • R. Dürichen, T. Wissel, F. Ernst, A. Schlaefer, A. Schweikard (2014). Multivariate respiratory motion prediction. Phys Med Biol. 59 (20), 6043-6060. [doi] [BibTex]

  • B. Stender, F. Ernst, B. Wang, Z. X. Zhang, A. Schlaefer (2013). Motion compensation of optical mapping signals from isolated beating rat hearts. SPIE. (Applications of Digital Image Processing XXXVI, SPIE), 88561C-1 - 88561C-6. [BibTex]

  • F. Ernst, R. Dürichen, A. Schlaefer, A. Schweikard (2013). Evaluating and comparing algorithms for respiratory motion prediction. Phys Med Biol. 58 (11), 3911-3929. [doi] [BibTex]

  • N. Lessmann, D. Drömann, A. Schlaefer (2013). Feasibility of respiratory motion compensated stereoscopic bronchoscopy. [BibTex]

  • R. Dürichen, O. Blanck, J. Dunst, G. Hildebrandt, A. Schlaefer, A. Schweikard (2013). Atemphasenabhängige Prädiktionsfehler in der extrakraniellen stereotaktischen Strahlentherapie. [BibTex]