Objective: The incidence of radiation pneumonitis (RP) has a highly linear relationship with low-dose lung volume. We previously established a volume-based algorithm (VBA) method to improve low-dose lung volume in radiotherapy (RT). This study assessed lung inflammatory changes by integrating fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (F-FDG PET/CT) with VBA for esophageal cancer patients undergoing arc-based RT. Methods: Thirty esophageal cancer patients received F-FDG PET/CT imaging pre-RT and post-RT were included in a retrospective pilot study. We fused lung doses and parameters of PET/CT in RT planning. Based on VBA, we used the 5Gy isodose curve to define high-dose (HD) and low-dose (LD) regions in the lung volume. We divided patients into non-RP (nRP) and RP groups. The maximum, mean standardized uptake value (SUVmax, SUVmean), global lung glycolysis (GLG), mean lung dose (MLD) and V in lungs were analyzed. Area under the curve values were utilized to identify optimal cut-off values for RP. Results: Eleven patients in the nRP group and 19 patients in the RP group were identified. In 30 RP lungs, post-RT SUVmax, SUVmean and GLG of HD regions showed significant increases compared to values for pre-RT lungs. There were no significant differences in values of 22 nRP lungs. Post-RT SUVmax and SUVmean of HD regions, MLD, and lung V and V in RP lungs were significantly higher than in nRP lungs. For detecting RP, the optimal cut-off values were post-RT SUVmax > 2.28 and lung V > 47.14%. Conclusion: This study successfully integrated F-FDG PET/CT with VBA to assess RP in esophageal cancer patients undergoing RT. Post-RT SUVmax > 2.28 and lung V > 47.14% might be potential indicators of RP.