Real-time motion-adaptive-optimization (MAO) in TomoTherapy

Weiguo Lu, Mingli Chen, Kenneth J. Ruchala, Quan Chen, Katja M. Langen, Patrick A. Kupelian, Gustavo H. Olivera

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

IMRT delivery follows a planned leaf sequence, which is optimized before treatment delivery. However, it is hard to model real-time variations, such as respiration, in the planning procedure. In this paper, we propose a negative feedback system of IMRT delivery that incorporates real-time optimization to account for intra-fraction motion. Specifically, we developed a feasible workflow of real-time motion-adaptive-optimization (MAO) for TomoTherapy delivery. TomoTherapy delivery is characterized by thousands of projections with a fast projection rate and ultra-fast binary leaf motion. The technique of MAO-guided delivery calculates (i) the motion-encoded dose that has been delivered up to any given projection during the delivery and (ii) the future dose that will be delivered based on the estimated motion probability and future fluence map. These two pieces of information are then used to optimize the leaf open time of the upcoming projection right before its delivery. It consists of several real-time procedures, including 'motion detection and prediction', 'delivered dose accumulation', 'future dose estimation' and 'projection optimization'. Real-time MAO requires that all procedures are executed in time less than the duration of a projection. We implemented and tested this technique using a TomoTherapy research system. The MAO calculation took about 100 ms per projection. We calculated and compared MAO-guided delivery with two other types of delivery, motion-without-compensation delivery (MD) and static delivery (SD), using simulated 1D cases, real TomoTherapy plans and the motion traces from clinical lung and prostate patients. The results showed that the proposed technique effectively compensated for motion errors of all test cases. Dose distributions and DVHs of MAO-guided delivery approached those of SD, for regular and irregular respiration with a peak-to-peak amplitude of 3 cm, and for medium and large prostate motions. The results conceptually proved that the proposed method is applicable for real-time motion compensation in TomoTherapy delivery. Extension of the method to real-time adaptive radiation therapy (ART) that compensates for all kinds of delivery errors was proposed. Further validation and clinical implementation is underway.

Original languageEnglish (US)
Pages (from-to)4373-4398
Number of pages26
JournalPhysics in Medicine and Biology
Volume54
Issue number14
DOIs
StatePublished - 2009

Fingerprint

Prostate
Respiration
Workflow
Radiotherapy
Lung
Research

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology

Cite this

Lu, W., Chen, M., Ruchala, K. J., Chen, Q., Langen, K. M., Kupelian, P. A., & Olivera, G. H. (2009). Real-time motion-adaptive-optimization (MAO) in TomoTherapy. Physics in Medicine and Biology, 54(14), 4373-4398. https://doi.org/10.1088/0031-9155/54/14/003

Real-time motion-adaptive-optimization (MAO) in TomoTherapy. / Lu, Weiguo; Chen, Mingli; Ruchala, Kenneth J.; Chen, Quan; Langen, Katja M.; Kupelian, Patrick A.; Olivera, Gustavo H.

In: Physics in Medicine and Biology, Vol. 54, No. 14, 2009, p. 4373-4398.

Research output: Contribution to journalArticle

Lu, W, Chen, M, Ruchala, KJ, Chen, Q, Langen, KM, Kupelian, PA & Olivera, GH 2009, 'Real-time motion-adaptive-optimization (MAO) in TomoTherapy', Physics in Medicine and Biology, vol. 54, no. 14, pp. 4373-4398. https://doi.org/10.1088/0031-9155/54/14/003
Lu, Weiguo ; Chen, Mingli ; Ruchala, Kenneth J. ; Chen, Quan ; Langen, Katja M. ; Kupelian, Patrick A. ; Olivera, Gustavo H. / Real-time motion-adaptive-optimization (MAO) in TomoTherapy. In: Physics in Medicine and Biology. 2009 ; Vol. 54, No. 14. pp. 4373-4398.
@article{1ac2d42821f74898b72b08185d4f0927,
title = "Real-time motion-adaptive-optimization (MAO) in TomoTherapy",
abstract = "IMRT delivery follows a planned leaf sequence, which is optimized before treatment delivery. However, it is hard to model real-time variations, such as respiration, in the planning procedure. In this paper, we propose a negative feedback system of IMRT delivery that incorporates real-time optimization to account for intra-fraction motion. Specifically, we developed a feasible workflow of real-time motion-adaptive-optimization (MAO) for TomoTherapy delivery. TomoTherapy delivery is characterized by thousands of projections with a fast projection rate and ultra-fast binary leaf motion. The technique of MAO-guided delivery calculates (i) the motion-encoded dose that has been delivered up to any given projection during the delivery and (ii) the future dose that will be delivered based on the estimated motion probability and future fluence map. These two pieces of information are then used to optimize the leaf open time of the upcoming projection right before its delivery. It consists of several real-time procedures, including 'motion detection and prediction', 'delivered dose accumulation', 'future dose estimation' and 'projection optimization'. Real-time MAO requires that all procedures are executed in time less than the duration of a projection. We implemented and tested this technique using a TomoTherapy research system. The MAO calculation took about 100 ms per projection. We calculated and compared MAO-guided delivery with two other types of delivery, motion-without-compensation delivery (MD) and static delivery (SD), using simulated 1D cases, real TomoTherapy plans and the motion traces from clinical lung and prostate patients. The results showed that the proposed technique effectively compensated for motion errors of all test cases. Dose distributions and DVHs of MAO-guided delivery approached those of SD, for regular and irregular respiration with a peak-to-peak amplitude of 3 cm, and for medium and large prostate motions. The results conceptually proved that the proposed method is applicable for real-time motion compensation in TomoTherapy delivery. Extension of the method to real-time adaptive radiation therapy (ART) that compensates for all kinds of delivery errors was proposed. Further validation and clinical implementation is underway.",
author = "Weiguo Lu and Mingli Chen and Ruchala, {Kenneth J.} and Quan Chen and Langen, {Katja M.} and Kupelian, {Patrick A.} and Olivera, {Gustavo H.}",
year = "2009",
doi = "10.1088/0031-9155/54/14/003",
language = "English (US)",
volume = "54",
pages = "4373--4398",
journal = "Physics in Medicine and Biology",
issn = "0031-9155",
publisher = "IOP Publishing Ltd.",
number = "14",

}

TY - JOUR

T1 - Real-time motion-adaptive-optimization (MAO) in TomoTherapy

AU - Lu, Weiguo

AU - Chen, Mingli

AU - Ruchala, Kenneth J.

AU - Chen, Quan

AU - Langen, Katja M.

AU - Kupelian, Patrick A.

AU - Olivera, Gustavo H.

PY - 2009

Y1 - 2009

N2 - IMRT delivery follows a planned leaf sequence, which is optimized before treatment delivery. However, it is hard to model real-time variations, such as respiration, in the planning procedure. In this paper, we propose a negative feedback system of IMRT delivery that incorporates real-time optimization to account for intra-fraction motion. Specifically, we developed a feasible workflow of real-time motion-adaptive-optimization (MAO) for TomoTherapy delivery. TomoTherapy delivery is characterized by thousands of projections with a fast projection rate and ultra-fast binary leaf motion. The technique of MAO-guided delivery calculates (i) the motion-encoded dose that has been delivered up to any given projection during the delivery and (ii) the future dose that will be delivered based on the estimated motion probability and future fluence map. These two pieces of information are then used to optimize the leaf open time of the upcoming projection right before its delivery. It consists of several real-time procedures, including 'motion detection and prediction', 'delivered dose accumulation', 'future dose estimation' and 'projection optimization'. Real-time MAO requires that all procedures are executed in time less than the duration of a projection. We implemented and tested this technique using a TomoTherapy research system. The MAO calculation took about 100 ms per projection. We calculated and compared MAO-guided delivery with two other types of delivery, motion-without-compensation delivery (MD) and static delivery (SD), using simulated 1D cases, real TomoTherapy plans and the motion traces from clinical lung and prostate patients. The results showed that the proposed technique effectively compensated for motion errors of all test cases. Dose distributions and DVHs of MAO-guided delivery approached those of SD, for regular and irregular respiration with a peak-to-peak amplitude of 3 cm, and for medium and large prostate motions. The results conceptually proved that the proposed method is applicable for real-time motion compensation in TomoTherapy delivery. Extension of the method to real-time adaptive radiation therapy (ART) that compensates for all kinds of delivery errors was proposed. Further validation and clinical implementation is underway.

AB - IMRT delivery follows a planned leaf sequence, which is optimized before treatment delivery. However, it is hard to model real-time variations, such as respiration, in the planning procedure. In this paper, we propose a negative feedback system of IMRT delivery that incorporates real-time optimization to account for intra-fraction motion. Specifically, we developed a feasible workflow of real-time motion-adaptive-optimization (MAO) for TomoTherapy delivery. TomoTherapy delivery is characterized by thousands of projections with a fast projection rate and ultra-fast binary leaf motion. The technique of MAO-guided delivery calculates (i) the motion-encoded dose that has been delivered up to any given projection during the delivery and (ii) the future dose that will be delivered based on the estimated motion probability and future fluence map. These two pieces of information are then used to optimize the leaf open time of the upcoming projection right before its delivery. It consists of several real-time procedures, including 'motion detection and prediction', 'delivered dose accumulation', 'future dose estimation' and 'projection optimization'. Real-time MAO requires that all procedures are executed in time less than the duration of a projection. We implemented and tested this technique using a TomoTherapy research system. The MAO calculation took about 100 ms per projection. We calculated and compared MAO-guided delivery with two other types of delivery, motion-without-compensation delivery (MD) and static delivery (SD), using simulated 1D cases, real TomoTherapy plans and the motion traces from clinical lung and prostate patients. The results showed that the proposed technique effectively compensated for motion errors of all test cases. Dose distributions and DVHs of MAO-guided delivery approached those of SD, for regular and irregular respiration with a peak-to-peak amplitude of 3 cm, and for medium and large prostate motions. The results conceptually proved that the proposed method is applicable for real-time motion compensation in TomoTherapy delivery. Extension of the method to real-time adaptive radiation therapy (ART) that compensates for all kinds of delivery errors was proposed. Further validation and clinical implementation is underway.

UR - http://www.scopus.com/inward/record.url?scp=70349577486&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349577486&partnerID=8YFLogxK

U2 - 10.1088/0031-9155/54/14/003

DO - 10.1088/0031-9155/54/14/003

M3 - Article

C2 - 19550000

AN - SCOPUS:70349577486

VL - 54

SP - 4373

EP - 4398

JO - Physics in Medicine and Biology

JF - Physics in Medicine and Biology

SN - 0031-9155

IS - 14

ER -