The Value Of MRI Radiomics Nomogram In Predicting The Grade Of Brain Edema After Meningioma Surgery

Objective: To explore the related factors of cerebral edema after meningioma surgery Method: Preoperative MRI images and postoperative CT images of 170 patients who were pathologically confirmed as meningioma by total surgical resection in our hospital from November 2019 to April 2023 were retrospectively collected,and relevant clinical data of patients with mild and severe postoperative cerebral edema were collected,and the relationship between these clinical factors and postoperative cerebral edema was statistically analyzed.Results: Among 170 patients,101 patients(59.4%)had mild postoperative cerebral edema,69 patients(40.6%)had severe cerebral edema,The 69 patients with severe cerebral edema were analyzed in terms of age,gender,whether they were complicated with hypertension,diabetes,preoperative intracranial hypertension,tumor size,tumor location,preoperative peritormal edema and tumor pathological classification.There was significant correlation between hypertension,tumor size,tumor location,preoperative peritumoral edema and pathological classification of tumor and severe postoperative cerebral edema.There were no significant differences in age,gender,diabetes mellitus and intracranial hypertension between the severe and mild cerebral edema groups.Conclusion: Postoperative severe cerebral edema in patients with meningioma is correlated with hypertension,tumor size,tumor location,preoperative peritumoral edema and pathological classification of tumor.Objective: To investigate the value of the imaging feature model based on T1 WI,T2WI and enhanced T1WI(T1WI+c)sequence images in predicting the severity of cerebral edema after meningioma surgery,so as to provide theoretical support for clinical follow-up treatment and improvement of prognosis.Methods: Clinical data,preoperative MRI images and postoperative CT images of 170 patients who were pathologically confirmed as meningioma by total surgical resection in our hospital from November 2019 to April 2023 were retrospectively collected,including 69 patients with mild cerebral edema and 101 patients with severe cerebral edema.A total of 170 patients were randomly divided into the training group(119 cases)and the verification group(51 cases)according to a ratio of 7:3.Image features of T1 WI,T2WI and T1WI+C were extracted using Darwin imaging platform.Two dimensional zones of interest(ROI)were manually delineated on three sequential images.According to the maximum absolute value normalization and selection operator regression,feature dimension reduction and optimal feature screening(number)were carried out.Finally,logistic regression was used to establish the image omics model.Then,8 clinical risk factors were included,and the better clinical risk factors were selected by multi-factor logistic regression,and the clinical prediction model was established.Finally,combined with imaging features and clinical risk factors,a clinical-imaging model was established.The efficacy of three models in predicting postoperative edema was evaluated by area under characteristic curve(ROC).Results: In the training group and the validation group,the AUCs of the imaging model,the clinical model and the clinical-imaging model were respectively,and the diagnostic efficacy of the clinical-imaging model was better than the other two models.Conclusion: The clinical-imaging map constructed based on the multi-sequence magnetic resonance imaging features and clinical features can provide certain clinical reference value for the classification of cerebral edema after meningioma surgery.