基于放射剂量及影像组学预测乳腺癌放疗患者放射性皮炎发生风险的分析OA

Objective To analyze the correlation between radiation dose parameters and radiomics in predicting the risk of radiation dermatitis in breast cancer patients undergoing radiotherapy.Methods A total of 285 breast cancer patients who re-ceived radiotherapy at our hospital from March 2021 to March 2024 were enrolled.All patients underwent radiation dose calculation and radiomics analysis before treatment.Patients were divided into radiation dermatitis group and control group according to the oc-currence of radiation dermatitis during radiotherapy.Clinical characteristics and imaging parameters were compared between the two groups.Multivariable Logistic regression was performed to identify independent factors associated with radiation dermatitis.Results Among 285 patients,3 cases dropped out,and among the remaining 282 cases,115 cases developed radiation dermatitis(the der-matitis group),with a radiation dermatitis incidence rate of 40.78%.There was no statistically significant difference in the general data and total radiation dose between the two groups of patients(P＞0.05).However,the planned target volume,average radiation dose and texture feature parameters in the dermatitis group were higher than those in the control group(P＜0.05).Morphological parameters-surface volume ratio and histogram parameter-uniformity in the dermatitis group were lower than those in the control group(P＜0.05).Multivariate Logistic regression analysis revealed that the planned target volume(OR=1.009),the average ra-diation dose of the target area(OR=1.201),and the texture feature parameter(the variance of the full-angle cluster)(OR=1.741)were independent risk factors for radiation dermatitis in breast cancer radiotherapy patients;the morphological parameter(the ratio of surface area)(OR=0.990)and the histogram parameter(the uniformity)(OR=0.014)were protective factors for radiation dermatitis in breast cancer radiotherapy patients.The C-index of the nomogram constructed for predicting radiation derma-titis in breast cancer patients undergoing radiotherapy was 0.926(95%CI 0.865～0.953).The calibration curve indicated a good correlation between the risk of radiation dermatitis predicted by the nomogram model and the internal sampling results.Conclusion The nomogram based on planned target volume,mean target dose,and all-angle cluster prominence variance can effectively pre-dict the risk of radiation dermatitis in breast cancer patients undergoing radiotherapy,with favorable clinical application value.