Encrypted login | home

Program Information

Evaluation of Different Knowledge-Based Planning Model for Esophageal Cancer

no image available
C Tao

C Tao1*, T Liu1 , J Lu1 , Y Yin1 , (1) Shandong Cancer Hospital Affiliated to Shandong University, Jinan, Shandong


SU-I-GPD-T-322 (Sunday, July 30, 2017) 3:00 PM - 6:00 PM Room: Exhibit Hall

Purpose: This study aimed to evaluate the planning solution of knowledge-based model with 3 different libraries for esophageal cancer. And the prediction accuracy of this model was accessed

Methods: 10 clinical patients with esophageal cancer (with supraclavicular lymph node) were enrolled as evaluation group in this study. Based on Rapidplan system (Varian), knowledge-based model, Model-30-s and Model-30-d, were generated including clinical IMRT plans of 30 IMRT plans for esophageal cancer with and without supraclavicular lymph node, respectively. Furthermore, the Model-60 was established by combining in these two models. These three knowledge-based models were applied to create knowledge-based plans for the evaluation group. All plans were set to the same prescription. And by considering total lung (V5, V10, V20 and Dmean), heart (V30, V40 and Dmean) and spinal-cord(Dmax) as the main organ at risk(OAR), the quality of these three knowledge-based plans were evaluated and compared.

Results: Compared with Model-30-s (V20 = 24.47, Dmean = 12.43 Gy), Model-60 model could produce lower V20 and Dmean value of total lung, with comparable PTV coverage (V20 = 23.8, Dmean = 12.28 Gy, p-values were 0.02 and 0.03 for V20 and Dmean, respectively). However, there were no significant difference observed at comparison between Model-30-s and Model-30-d in all endpoints.

Conclusion: By involving more treatment plans, the knowledge-based model was more accurate for planning. However, with or without supraclavicular lymph node did not significantly effect the planning solution.

Contact Email: