An Efficient Compression Technique using Modified Wavelet Transform for Brain Tumor Images

With the modern advancement in communication and E-Medical technologies which leads to an important responsibility in telemedicine as this helps health center in acquiring patient information even when they are away from patients. Also the need processing medical image in digital for diagnosis has been significantly increased as a result of digital radiology. Thus this imposes great challenges in storing and transmitting medical images without any degradation. The proposed methodology is a lossless compression scheme specific for Medical Resonance Imaging (MRI) brain images containing tumor. It involves the partition of the input image into region of importance (ROI) and region of non importance (RONI) using K-means algorithm. Differential Pulse Code Modulation (DPCM) with Curvelet transform is applied for the ROI part. Discrete Wavelet Transform (DWT) followed by Set Partitioning in Hierarchical Tree (SPIHT) and adaptive arithmetic coding is applied for RONI part. The combined compressed outputs result in the final compressed image, and the performance metrics, including peak signal-to-noise ratio, compression ratio, and entropy, demonstrate that the proposed method outperforms existing techniques.