Signals @ ROB

Tools

Parser for CyberKnife log files
- MATLAB MEX files [win32 | win64 | linux_i386 | linux_x64]
- Binaries: [win32 | win64 | linux_i386 | linux_x64]
The Prediction Toolkit is available for download here:
- Windows binaries here
- Linux binaries here
- Sources here
Visual C++ runtimes [win32 | win64]

If you use the Prediction Toolkit for one of your publications, please cite one of the following:

Norman Rzezovski and Floris Ernst, Graphical Tool for the Prediction of Respiratory Motion Signals, in: 7. Jahrestagung der Deutschen Gesellschaft für Computer- und Roboterassistierte Chirurgie, CURAC, pages 179--180, 2008, see here
Floris Ernst, Compensating for Quasi-periodic Motion in Robotic Radiosurgery, Springer, 2011, DOI 10.1007/978-1-4614-1912-9, see here

If you use these scripts for one of your publications, please cite one of the following:

Floris Ernst, Ralf Bruder, Alexander Schlaefer and Achim Schweikard, Performance Measures and Pre‐Processing for Respiratory Motion Prediction, in: 53rd Annual Meeting of the AAPM, pages 3857, 2011, DOI 10.1118/1.3613523, see here
Floris Ernst, Compensating for Quasi-periodic Motion in Robotic Radiosurgery, Springer, 2011, DOI 10.1007/978-1-4614-1912-9, see here

If you use the RVM prediction script for one of your publications, please cite one of the following:

R. Dürichen, T. Wissel, F. Ernst, A. Schweikard: “Respiratory Motion Compensation with Relevance Vector Machines”, MICCAI, Lecture Notes in Computer Science Volume 8150, pp 108-115, 2013, DOI 978-3-642-40763-5_14, see here

Respiratory motion traces recorded during CyberKnife treatment at Georgetown University Hospital. Data by courtesy of Dr. Kevin Cleary and Dr. Sonja Dieterich.
- Original data (unprocessed, 334 MB, 306 traces)
- Processed data (cut & PCA, 202 MB, 304 traces)

Bi-modal respiratory motion traces. Data by courtesy of Floris Ernst.
- MATLAB data, seven human liver motion traces (4D ultrasound, 17.5 to 21.3 Hz), acquired with template matching, external data from IR LED, 5-6 minutes long
- MATLAB data, two porcine liver motion traces (biplanar fluoroscopy, triangulated, 15 Hz), external data from IR LEDs

Cardiac motion traces. Data by courtesy of Floris Ernst.
- Externally measured data (optical tracking of the apex beat, 16 traces)
- Internal motion (Volumetric ultrasound, tracked with template matching, 10 traces)
- Bi-modal external data (simultaneous optical tracking of the apex beat and acquisition of an ECG signal, two traces)

If you use this data for one of your publications, please cite:

Floris Ernst, Compensating for Quasi-periodic Motion in Robotic Radiosurgery, Springer, 2011, DOI 10.1007/978-1-4614-1912-9, see here

Multivariate motion traces. Data by courtesy of Robert Dürichen.
- Multivariate external artefact data (optical, flow, strain and acceleration data, 18 traces, 4-5 minutes long, each traces contains 10 motion artefacts (e.g. coughing, yawning), sampling frequency 26 Hz)
- Multivariate external long term data (optical, flow, strain and acceleration data, 18 traces, 19-20 minutes long, unconditioned normal breathing, sampling frequency 26 Hz)
- Multivariate external and internal long term data (optical, flow, strain, acceleration and ultrasound data, 7 traces, 15-20 minutes long, unconditioned normal breathing, sampling frequency 17 Hz)

If you use this data for one of your publications, please cite:

Robert Dürichen et al., Evaluation of the potential of multi-modal sensors for respiratory motion prediction and correlation , International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC’13), Osaka, Japan, 2013, DOI EMBC.2013.6610839, see here

Consult the User's Guide for information on using the wiki software.