摘要：This study aimed to investigate the impact of artificial intelligence-assisted intensity-modulated radiotherapy (IMRT) plan optimization on the radiation doses received by the rectum and bladder as well as radiation-induced injuries in patients with locally advanced cervical cancer. A total of 100 patients with locally advanced cervical cancer were enrolled and divided into a conventional IMRT group and an artificial intelligence-assisted IMRT group. The results showed that in terms of the radiation doses to the rectum and bladder, all dosimetric parameters (such as mean dose, maximum dose, and volume-dose parameters, etc.) in the artificial intelligence-assisted group were significantly lower than those in the conventional group (p < 0.05). Regarding radiation-induced injuries, the incidences and severities of both acute and late radiation-induced proctitis and
cystitis in the artificial intelligence-assisted group were lower than those in the conventional group (p < 0.05). These findings suggest that artificial intelligence-assisted IMRT plan optimization can effectively reduce the radiation doses to the rectum and bladder and decrease radiation-induced injuries in patients with locally advanced cervical cancer, which is expected to provide a more precise and safer treatment strategy for radiotherapy of locally advanced cervical cancer. However, this study has limitations such as a limited sample size and being a single-center study. Future research with multi-center, large-sample, and more in-depth investigations is needed.