Click here to

Session: Therapy General ePoster Viewing [Return to Session]

Development and Validation of Knowledge Based Planning Model for Prostate Node Positive Stereotactic Body Radiation Therapy

R Phurailatpam1*, D Patil2, V Raveendran3, K Joshi4, J Pk5, P Maitre6, V Murthy7, (1) Actrec, Tata Memorial Centre, Maharastra, ,IN, (2) ACTREC, Ratnagiri, ,IN, (3) Advanced Center for Treatment, Research and Education in Cancer (ACTREC), Tata Memorial Center, Navi Mumbai, MH, IN, (4) TATA MEMORIAL CENTRE, ACTREC, NAVI MUMBAI, Navi Mumbai, ,IN, (5) Actrec,tmc, ,,(6) Actrec,tmc, ,,(7) Actrec,tmc,


PO-GePV-T-314 (Sunday, 7/10/2022)   [Eastern Time (GMT-4)]

ePoster Forums

Purpose: Development of Knowledge-based planning (KBP) model for prostate node-positive stereotactic body radiation therapy.

Methods: The KBP model was created by Varian Rapid PlanTM (V.16.1) using sixty clinical patient plans 36.35 Gy was prescribed for the prostate planning target volume (PTV) and 25 Gy for the nodal region (PTV 25 Gy). Volumetric modulated arc therapy (VMAT) plans were generated with two partial arcs (250°-110°) of 10MV flattening filter-free beam. For validation, three types of plans were generated for fifteen patients: 1. KBP without manual interruption (oKBP) in single optimization with intermediate-dose option and convergence mode enabled. 2. KBP with manual interruption (mKBP) to control spill of 95% of 25Gy . 3 Manually optimized plans. Plans 1 and 2 were compared with plan 3 in terms of target coverage, Paddick conformity index (CI), homogeneity index (HI), and OARs sparing. Statistical analysis was done using t-test or Wilcoxon signed-rank test. The complexity metric (CM) of plans was computed using in house script. Plan deliverability was checked by patient-specific QA (PSQA).

Results: No significant difference was observed between the plans for target coverage (p=0.460) nor target CI (p=0.966), however oKBP and mKBP plans showed statistically significant differences in HI for both the targets (p= 0.003, p=0.001 respectively). oKBP and mKBP plans showed better OAR sparing. oKBP plans showed least complexity with average CM of 0.1370 compared to clinical plans (CMavg= 0.1654, p= 0.01) and mKBP plan (CMavg= 0.1511, p= 0.65). The increased CM for manual plans reflected in the delivery efficiency as KBP plans showed lesser gamma failures (p=0.047).

Conclusion: KBP model for the prostate node-positive SBRT was created and validated. The KBP model can achieve consistent quality plans in lesser planning time than the manual plan, with lesser complexity and better delivery efficiency.


Validation, Radiation Therapy


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

Contact Email