IntOPMICM: Intelligent Medical Image Size Reduction Model.

Area of Science:

• Medical Imaging
• Image Compression
• Computational Intelligence

Background:

• Increasing use of medical images necessitates efficient compression.
• Lossy compression methods face challenges in balancing compression ratio and image quality.
• Existing techniques often struggle to maintain diagnostic accuracy.

Purpose of the Study:

• To introduce the IntOPMICM technique for enhanced medical image compression.
• To evaluate the performance of IntOPMICM against existing methods.
• To improve the compression ratio and image quality metrics for medical imaging.

Main Methods:

• Developed IntOPMICM, a hybrid scheme combining Genetic Particle Swarm Optimization (GenPSO) and Vector Quantization (VQ).
• Utilized a codebook generated through a combination of fragments and genetic algorithms.
• Evaluated performance using metrics like Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), Structural Similarity Index Measure (SSIM), Normalized Mean Squared Error (NMSE), Signal-to-Noise Ratio (SNR), and Compression Ratio (CR).

Main Results:

• IntOPMICM demonstrated higher PSNR and SSIM values at given compression ratios compared to existing methods.
• The technique achieved lower MSE, Root Mean Squared Error (RMSE), and SNR values for comparable compression ratios.
• Experimental results on real-time medical imaging validated the proposed approach's effectiveness.

Conclusions:

• The IntOPMICM technique offers a significant improvement in medical image compression.
• It provides a better balance between compression efficiency and diagnostic image quality.
• IntOPMICM represents a promising advancement for medical imaging applications.