Click here to


Are you sure ?

Yes, do it No, cancel

A Robust Index for Determining the Optimal Direction of Target Matching in IGRT

H Wang*, Y Zhou , X Wang , Y Zhang , I Vergalasova , N Yue , Z Xu , K Nie , Rutgers Cancer Institute of New Jersey, New Brunswick, NJ


(Tuesday, 7/16/2019) 10:00 AM - 10:30 AM

Room: Exhibit Hall | Forum 9

Purpose: IGRT requires target matching at treatment. However, it is often that a target exhibits volumetric and shape changes at treatment from simulation and exact matching is impossible. The question is whether there exists an optimal direction along which satisfactory target matching yields the best possible dose coverage. This study is to develop a robust index for determining the optimal direction.

Methods: A methodology was developed to compute the distances between the surface of a target volume and various dose surfaces along various directions. An index was developed and computed to determine the optimal direction for 2 lung cancer cases(One case was prescribed for 2Gyx30f;the other case was prescribed for 10Gyx5f).The distances of d(100%p-CTV)2Gy or d(100%p-ITV)10Gy and d(90%p-100%p) along the eighteen directions (SI, LR, AP and the 45o angles between them) were reviewed across the center of tumor. Pearson correlation analyses between d(100%p-CTV)2Gy or d(100%p-ITV)10Gy and d(90%p-100%p) were completed , and a robust index DSI based on dsecurity was presented as: DSI(i)=Ndsecurity(i)(-1)/Σdsecurity(i)(-1) and DSI10Gy(i)=2Ndsecurity(i)(-1)/Σdsecurity(i)(-1) (N: Quantity of evaluated directions, i∈(1~N)).

Results: Mean value and standard deviation of d(100%p-CTV)2Gy and d(90%p-100%p)2Gy for case 1 were 1.33±0.74cm and 0.63±0.36cm, while mean value and standard deviation of d(100%p-ITV)10Gy and d(90%p-100%p)10Gy for case 2 were 0.64±0.17cm and 0.24±0.08cm. As there was no linear correlation between d(100%p-CTV)2Gy or d(100%p-ITV)10Gy and d(90%p-100%p)(r=0.122, p=0.631 and r=0.027, p=0.915). Mean value and standard deviation of DSI2Gy and DSI10Gy for two cases were 1.00±0.503 and 2.00±0.406. Based on the index, it was determined that d(90%p-CTV)2Gy and d(90%p-ITV)10Gy were vital parameters related to the target coverage, for the two cases, and the optimal directions were then determined correspondingly.

Conclusion: A robust index is developed for the determination of the optimal direction of target matching in IGRT. Its applicability and prediction accuracy will be evaluated in a larger pool of cases.


Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email