The Complexity Of Multi-scale Texture Analysis Of Medical Images

In the past 30 years , texture analysis is one of hot research subjects within the fields of computer vision , visual physiology and psychophysis . Currently, methods of image texture analysis are undergoing great development and utilization in fields of medical imaging. Given the general interest and striking growth in computer-aided diagnosis (CAD), the application of texture analysis in the diagnostic interpretation of radiologic image has become a rapidly expanding field of research.The texture parameter is different between normal and abnormal tissue CT image. If we can use medical image processing to discrimination them, it's a meaningful work for medical assist diagnosis. This paper investigated Multi-Scale Complexity algorithm and evaluated it's performance in differentiating normal liver CT image and Primary Carcinoma of the Liver CT image. Due to Primary hepatocellular Carcinoma of the Liver CT images appear to change in some scale of texture parameter, not whole, we gives a Multi-Scale Complexity algorithm to analysis them. Through clinic data experiment, this measurement achieved high discrimination ratio .The promising results demonstrate Multi-Scale Complexity measurements potential in liver CT image discrimination.The two-dimensional picture is different from the one-dimensional signal, This demands to improve the already existing complexity algorithm further. We applied multi-scale complexity analysis that suited for image texture analysis. We detailed depicted the problem of sequence and resolution factor using this method in image analysis and also studied the impact of rotated image and noise on result of the algorithm.