CT and MRI Image Diagnosis of Cystic Renal Cell Carcinoma Based on a Fractional-Order Differential Texture Enhancement Algorithm

摘要

Objective: Diagnosis and analysis of cystic renal cell carcinoma by CT and MRI images is performed based on a fractional-order differential texture enhancement algorithm. In our paper, we implemented three methods (histogram equalization, fractional differentials, and proposed fractional differentials based on wavelet transform) to perform the texture enhancement experiments and compared their performance. Methods: We selected 30 patients with cystic renal cell carcinoma. The characteristics of CT and MRI imaging were summarized and analyzed by texture enhancement algorithm, and compared with pathological diagnosis results. Results: Comparison of different texture enhancement algorithm can be concluded that the differential algorithm based on wavelet transform has obvious effect. In the images processed by the texture enhancement algorithm, we can determine that most patients do not have typical symptoms. By performing CT and MRI enhanced scans, irregular separation, wall nodules, and calcification can be seen. CT and MRI imaging and pathology. The results were compared (P > 0.05), and the MRI lesion details showed a higher coincidence rate with CT (P < 0.05). Conclusion: After processing with the order differential algorithm based on wavelet transform, the texture details can be effectively preserved and the misdiagnosis rate of cystic renal cell carcinoma can be reduced. Besides, the diagnostic coincidence rate of MRI is higher than that of CT. This image processing technique can faciliate the diagnosis of cystic renal cell carcinoma effectively.