Video compression of coronary angiograms based on discrete wavelet transform with block classification

Area of Science:

• Medical imaging
• Image processing
• Video compression

Background:

• Angiographic images require high compression for efficient storage and transmission.
• Existing video compression methods like MPEG can introduce block artifacts, compromising diagnostic accuracy.
• Loss of diagnostic information due to compression artifacts is a critical concern in medical imaging.

Purpose of the Study:

• To develop a new video compression method for angiographic images.
• To achieve a high compression ratio (~20:1) while eliminating block artifacts.
• To maintain or improve diagnostic accuracy in compressed angiographic sequences.

Main Methods:

• The method utilizes motion compensated prediction from Motion Picture Experts Group (MPEG) standards.
• It employs Discrete Wavelet Transform (DWT) for encoding error images, unlike the block Discrete Cosine Transform (DCT) in MPEG.
• A novel classification scheme labels image blocks as intra, error, or background for tailored encoding.

Main Results:

• The developed method achieves a compression ratio of approximately 20:1.
• Block artifacts, which degrade diagnostic accuracy, are effectively eliminated.
• The hybrid coding approach demonstrates improved compression efficiency in specific scenarios.

Conclusions:

• This new video compression technique offers a viable solution for medical imaging applications.
• The combination of DWT and a block classification scheme successfully addresses the limitations of traditional methods.
• The method's generalizability extends to other dynamic image sequence applications sensitive to block artifacts.