Medical Image Segmentation Method Based on Information Fusion

A new method for medical image segmentation with adequate texture information is proposed, in which texture and gray variation information were integrated. A pyramid data structure-based region growing algorithm was designed and applied to medical image segmentation. In this algorithm, the combination of texture information and gray variation information was implemented inside the region. Meanwhile, the former was mainly used near the edge of region. The experimental result shows that this method can produce good segmentation result when applied to certain medical images.