Controlling motion prediction errors in radiotherapy with relevance vector machine

Type of publication:  Artikel
Zeitschrift: International Journal of Computer Assisted Radiology and Surgery
Jahr: 2014
Seiten: in press
Notiz: Motion Compensation in Radiosurgery
DOI: 10.1007/s11548-014-1008-x
Abriss: Purpose: Robotic radiotherapy can precisely ablate moving tumors when time latencies have been compensated. Recently, relevance vector machines (RVM), a probabilistic regression technique, outperformed six other prediction algorithms for respiratory compensation. The method has the distinct advantage that each predicted point is assumed to be drawn from a normal distribution. Second order statistics, the predicted variance, were used to control RVM prediction error during a treatment and to construct hybrid prediction algorithms. Methods: First, the duty cycle and the precision were correlated to the variance by interrupting the treatment if the variance exceeds a threshold. Second, two hybrid algorithms based on the variance were developed, one consisting of multiple RVMs (HYB R V M) and the other of a combination between a wavelet based least mean square algorithm (wLMS) and a RVM (HYB w LM S − R V M). The variance for different motion traces was analyzed to reveal a characteristic variance pattern which gives insight in what kind of prediction errors can be controlled by the variance. Results: Limiting the variance by a threshold resulted in an increased precision with a decreased duty cycle. All hybrid algorithms showed an increased prediction accuracy compared to using only their individual algorithms. The best hybrid algorithm, HYB R V M, can decrease the mean RMSE over all 304 motion traces from 0 .18 mm for a linear RVM to 0 .17 mm. Conclusions: The predicted variance was shown to be an efficient metric to control prediction errors, resulting in a more robust radiotherapy treatment. The hybrid algorithm HYB R V M could be translated to clinical practice. It do es not require further parameters, can b e completely parallelised and easily further extended.
Schlagworte: Computer-assisted radiation therapy, Relevance vector machines, Respiratory motion compensation, Uncertainty measures
Autoren: Dürichen, Robert
Wissel, Tobias
Schweikard, Achim
Anhänge
  • 2014 - JCARS - Dürichen - Cont...
 [Bib|RIS]