【摘要】 目的　提高13 1I治疗剂量的合理性。方法　量化13 1I治疗Graves病的多种相关因素 ,将2 0 6例确诊为Graves病的患者进行分组 ,其中 1 0 9例行神经网络计算13 1I剂量法进行治疗 ,另 97例以传统的单位甲状腺质量吸收剂量法治疗 ,并进行比较。结果　神经网络计算13 1I剂量法治疗组与单位甲状腺质量吸收剂量法治疗组的有效率分别为 93.5%和 90 .7% ,治愈率分别为 90 .8%和 85.5% ,复发率分别为 3.6 %和 5.1 % ,早发性甲状腺功能减低 (简称甲低 )发生率分别为 3.6 %和 5.1 %。结论　应用人工神经网络计算13 1I剂量治疗Graves病可行 ,可规范和指导13 1I治疗Graves病 ,减少复发和早发性甲低。更多还原

【Abstract】 Objective To study the feasibility of radioactive treatment of Graves disease with dosimetery of ~ 131I in novel method by computer. Methods The database comprised the 206 selected Graves disease cases and according to this database the nonlinear functions could be inferred consequently, then 206 Graves disease patients were divided into two groups. The group one, 109 patients, was treated with back-propagation (BP) dosimetery method. The group two was treated with traditional method. The results were analyzed. Results Group one had an effective rate of 93.5%, cure rate of 90.8%, recurrent rate of 3.6% and early stage of hypothyroidism rate of 3.6%. Group two had an effective rate of 90.7%, cure rate of 85.5%,recurrent rate of 5.1% and early stage of hypothyroidism rate of 5.1%. Conclusion The BP method is efficient and the results reflect that this method might increase the accuracy of the dosimetery of ~ 131I and reduce recurrent rate and early stage of hypothyroidism occurrence.更多还原