MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel
  1. Home
  2. Meeting Program
  3. All
  4. The Research of Automatic Treatment Planning by DVH Prediction Based On Classifying the OARs Overlap

The Research of Automatic Treatment Planning by DVH Prediction Based On Classifying the OARs Overlap

X Pan1,2*, D Wang1,3,4, J Jia1,3,4, L Hu1, FDS Team (1) Institute Of Nuclear Energy Safety Technology, Chinese Academy Of Sciences, CN (2) University of Science and Technology of China, Hefei Anhui, CN, (3) Anhui Engineering Technology Research Center of Accurate Radiotherapy, Hefei Anhui, CN (4) CAS SuperAccuracy Science & Technology Co., Ltd., Nanjing, Jiangsu, CN

Presentations

(Sunday, 7/12/2020)   [Eastern Time (GMT-4)]

Room: AAPM ePoster Library

Purpose:
Intensity-modulated Radiation Therapy (IMRT) has gradually become the standard treatment modality of Radiotherapy, and the quality of IMRT plan is largely dependent on the experience and time spent by medical physicist. Several studies were available on the review of automatic planning by dose volume histogram (DVH) prediction, but the overlap degree of planning target volume (PTV) and organs at risk (OARs) had not been analyzed. The automatic planning method by the DVH prediction using Pinnacle script was developed.

Methods:
Based on the anatomic structure of prostate cancer patients, the data was divided into a training set T?i? (i =1, 2, ...,6) and a test set B?i? (i =1, 2, 3). The PTV was expanded to cover OAR and then OAR was divided into several sub blocks. An accurate mathematical model that based on partial normal Gaussian function to predict the OARs sub-block DVH information was established. The relationship between the overlap degree and the sub block differential DVH was analyzed. Then the results of differential DVH were taken as the automatic dose constrains used for designing plan. The data of related evaluation criteria and the difference between the evaluation criteria of automatic treatment plans and the evaluation criteria of original plans was calculated.

Results:
The automatic method was tested by several manual planning. The training set’s result shown that, the overlap was greater, the result was less obvious. But the test set’s improvements in cases’ evaluation criteria were different. For case B?1?, case B?2?, case B?3?, their overlaps were 7.69%, 13.77%, and 34.72%. Most of the evaluation results of each case were improved by more than 1.6%, 1.35%, 0.1%.

Conclusion: automatic treatment planning method based on the predicted of OAR DVH based on classification of OAR overlap was developed and the OAR overlap could classify the plan quality.

Funding Support, Disclosures, and Conflict of Interest: the project of Anhui province (18030801135) National Natural Science of China (11605233)

Keywords

Dose Volume Histograms, Prostate Therapy, Treatment Planning

Taxonomy

TH- External Beam- Photons: Treatment planning using machine learning/Knowledge Based Planning/automation

Contact Email

Additional Links