ESTRO 2024 - Abstract Book

S4503

Physics - Machine learning models and clinical applications

ESTRO 2024

all 20 patients (table 1, figure 1). Statistically significant improvements included increased CTVelective dose(V94 sDL1: 98.5%, V94 sDL2: 98.8%, p <0.001), higher maximum body dose (D1cm3 sDL1: 74.1Gy, D1cm3 sDL2: 74.7Gy, p <0.001) and a notable average esophageal dose reduction of 2.7 Gy (p <0.001). These updates were implemented while maintaining comparable CTVprimary coverage (ΔV94 = 0.1 Gy, p = 0.022) and minimal differences in dysphagia NTCP (ΔNTCP = 0.2 Gy, p <0.001) and xerostomia NTCP values (ΔNTCP = 0.4 Gy, p <0.001), between sDL1 and sDL2 plans.