HYBRID PARTICLE SWARM OPTIMIZATION WITH WAVELET MUTATION BASED SEGMENTATION AND PROGRESSIVE TRANSMISSION TECHNIQUE FOR MRI IMAGES

A Hybrid particle swarm algorithm that incorporates a wavelet theory based mutation operation is used for segmentation and progressive transmission for Magnetic Resonance Images. The concept of irrelevancy reduction is used to reduce the load on the transmission network. We use Entropy maximization using Hybrid particle swarm algorithm with Wavelet based mutation operation to get the region of interest of the Magnetic Resonance Image. It applies the Multi-resolution Wavelet theory to enhance the Particle Swarm Optimization in exploring the solution space more effectively for a better solution. A variable rectangular mask is used to reduce the noise in the segmented image. Precision and Recall is used to evaluate segmentation accuracy. We use Discrete Cosine Transform to make progressive transmission system very efficient. Twin Distribution Entropy Coding is employed to compress the DCT coefficients. We use varying percentages of DCT coefficients of segmented MRI image for progressive image transmission. Clustering of the segmented MRI image using k-means algorithm results in predominantly two clusters namely that of lesions and background. At the receiver's end the doctor or a radiologist identifies a particular class of lesions and may ask for the entire un-segmented Magnetic Resonance Image dataset of a particular patient. Irrevalency reduction thus helps to reduce the load on the system by choosing not to send the images without lesions.