Medical Image Compression Discussed at CIU

At a seminar organized by the Faculty of Engineering at Cyprus International University (CIU), Dr. Ziya Arnavut, a faculty member at the State University of New York, shared the latest approaches in medical image compression technologies through his presentation titled “Rigorous PSNR and SSIM Error Bounds to Guarantee Image Quality in Medical Compression.” The seminar focused on how diagnostic reliability can be preserved while enabling more efficient storage and transmission of medical images.
During his presentation, Dr. Arnavut emphasized that although lossy compression methods provide significant storage savings, the most critical factor remains the preservation of image quality. He explained that their research establishes mathematical error bounds for PSNR and SSIM values, thereby providing assurance of image quality after compression.
Dr. Arnavut noted that the U.S. Food and Drug Administration (FDA) has highlighted the absence of a common standard for acceptable information loss in medical images. As a result, reliable metrics that evaluate image quality rather than compression ratios have become increasingly important. He stated that their newly developed algorithm can achieve near-lossless compression for grayscale medical images while mathematically guaranteeing minimum and average PSNR and SSIM values.
The seminar also featured comparative performance results between the proposed method and existing compression techniques. Dr. Arnavut presented experimental findings demonstrating that the algorithm offers significant advantages in data storage and transmission while maintaining image quality. During the question-and-answer session at the end of the seminar, participants exchanged views on how image compression technologies and AI-assisted diagnostic systems could be integrated and utilized together in the future.