A quantitative study to predict the risk of recurrence after radiotherapy for cervical cancer based on multi-sequence MRI imagingomics

Background: With the increasing clinical application of radiomics,the ability of magnetic resonance imaging omics to diagnose and predict the prognosis of diseases has attracted extensive attention in the scientific research community, and its application in the diagnosis, staging and treatment efficacy evaluation of cervical cancer has been gradually got attention.. Methods: All 52 patients in the radiotherapy group were cervical cancer patients and were diagnosed and treated with radiotherapy at our hospital from July 2019 to December 2021, while the healthy control group was selected from 41 healthy patients. The MRI performance of the two groups and the MRI performance of the patients in the radiotherapy group before and after treatment were observed separately. In addition, their ADC values were compared and the changes in ADC values before and after treatment were compared in patients with different prognosis in the radiotherapy group, compared the texture analysis parameters of patients with and without postoperative recurrence in radiotherapy group, and analyzed the predictive efficacy of MRI texture analysis parameters on the recurrence of cervical cancer patients after radiotherapy by ROC. Results: The inorganic mass and DWI values of patients with cervical cancer increased with their grading and ADC decreased with them; Comparing the ADC values before and after radiotherapy, it was found that the values in the radiotherapy group were obviously lower than those in the healthy group (P < 0.05); a total of 16 patients (30.77%) received radiotherapy, they had recurrence within 6 months after radiotherapy and 36 patients (69.23%) had no recurrence in the radiotherapy group; according to the recurrence of patients after radiotherapy. The ROC curve analysis showed that the area under the curve of entropy value was the largest, which confirmed its diagnostic effectiveness. Conclusion: Multi-series MRI imaging omics is of high clinical value in predicting the risk of recurrence after radiotherapy for cervical cancer patients.