RESUMO
OBJECTIVE: To review our single-institution experience in the surgical management of complex skull base tumors using multimodal image fusion technology. METHODS: From October 2019 to January 2022, 7 cases of complex skull base tumors that performed preoperative multimodal image fusion in Zhuhai People's Hospital neurosurgery department were involved in this study. The image data were uploaded to the GE AW workstation. Corresponding image sequences were opened in the workstation to complete registration fusion and 3D reconstruction. We retrospectively reviewed the clinical and imaging data, and surgical strategy, respectively. RESULTS: one case of recurrent C2 schwannoma, 1 case of recurrent spindle cell tumor of the left cranio-orbital communication, 1 case of lobular malignant tumor of the left infratemporal fossa, 1 case of central giant cell repairing granuloma, 1 case of mesenchymal malignant tumor in left pharyngeal process, 1 case of meningioma in jugular foramen, and 1 case of hemangioblastoma with vascular malformation in fourth ventricular. All cases underwent preoperative multimodal image fusion for the surgical plan and all cases had gross total resection. Except for one case of mesenchymal malignant tumor in left pharyngeal process that had dysphagia and one case of hemangioblastoma that had discoordination, others cases were without postoperative complication. CONCLUSIONS: Preoperative multimodal image fusion and surgical approach simulation benefit complex skull base tumor surgical treatment. Individually multiple image assessment of complex skull base tumors to determine the specific surgical strategy is more rational and should be recommended (Supplemental Digital Content 1, Supplementary Video, http://links.lww.com/SCS/F936).
RESUMO
OBJECTIVE: The conventional breast Diffusion-weighted imaging (DWI) was subtly influenced by microcirculation owing to the insufficient selection of the b values. However, the multiparameter derived from multiple b-value exhibits more reliable image quality and maximize the diagnostic accuracy. We aim to evaluate the diagnostic performance of stand-alone parameter or in combination with multiparameter derived from multiple b-value DWI in differentiating malignant from benign breast lesions. METHODS: A total of forty-one patients diagnosed with benign breast tumor and thirty-eight patients with malignant breast tumor underwent DWI using thirteen b values and other MRI functional sequence at 3.0 T magnetic resonance. Data were accepted mono-exponential, bi-exponential, stretched-exponential, aquaporins (AQP) model analysis. A receiver operating characteristic curve (ROC) was used to evaluate the diagnostic performance of quantitative parameter or multiparametric combination. The Youden index, sensitivity and specificity were used to assess the optimal diagnostic model. T-test, logistic regression analysis, and Z-test were used. P value < 0.05 was considered statistically significant. RESULT: The ADCavg, ADCmax, f, and α value of the malignant group were lower than the benign group, while the ADCfast value was higher instead. The ADCmin, ADCslow, DDC and ADCAQP showed no statistical significance. The combination (ADCavg-ADCfast) yielded the largest area under curve (AUC = 0.807) with sensitivity (68.42%), specificity (87.8%) and highest Youden index, indicating that multiparametric combination (ADCavg-ADCfast) was validated to be a useful model in differentiating the benign from breast malignant lesion. CONCLUSION: The current study based on the multiple b-value diffusion model demonstrated quantitatively multiparametric combination (ADCavg-ADCfast) exhibited the optimal diagnostic efficacy to differentiate malignant from benign breast lesions, suggesting that multiparameter would be a promising non-invasiveness to diagnose breast lesions.