Program Information
Computational Analysis for Dose Distributions of Cold Spots in Intensity Modulated Radiotherapy for Prostate Cancer
R Asamura1*, H Arimura2 , M Soufi1 , S Ohga3 , H Honda3 , T Sasaki2 , (1) Graduate School of Medical Sciences, Kyushu University, Japan , (2) Kyushu University, Japan , (4) Kyushu University Hospital, Japan
Presentations
SU-I-GPD-T-382 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose: To computationally analyze cold spots in dose distributions of intensity modulated radiotherapy (IMRT) plans for prostate cancer.
Methods: We have developed a computational tool for detection of cold spots in dose distributions obtained from IMRT. Eighteen cases of IMRT were used for the development of the proposed approach. In order to obtain PTV dose images, PTV images were interpolated by using a nearest neighbor interpolation method so as to fit the pixel value of PTV images with one of original dose images, because the pixel sizes of the two kinds of images are different. Cold spot images were obtained to apply thresholding to PTV dose images based on the definition of a cold spot. A cold spot is defined as a region with doses lower than 93% of a prescribed dose in PTV. From the above steps, the detection of cold spots was achieved. In addition, by applying 26-connected labeling to cold spot images, each cold spot region could be recognized, and then, the number and volume of cold spots were calculated.
Results: 127 cold spots with volumes less than 0.125 cm³ were detected with a mean number of 9.4 per case and a mean volume of 0.46 cm³ for the cold spots. Furthermore, there existed 12 cold spots with volume more than 1.25 cm³.
Conclusion: This study revealed that the cold spots in dose distributions of IMRT plans could not be negligible.
Contact Email: