Coding of echocardiographic image by selection of the normalization matrix using fuzzy logic

Area of Science:

• Medical Imaging
• Signal Processing
• Computer Science

Background:

• Echocardiographic images are crucial for cardiac diagnosis but require efficient storage and transmission.
• Existing compression algorithms may not optimally preserve the diagnostic quality of echocardiographic data.
• Adaptive compression techniques are needed to balance file size and image fidelity.

Purpose of the Study:

• To develop and evaluate a new algorithm for compressing echocardiographic images.
• To improve the quality of compressed echocardiographic images by adapting to local image characteristics.
• To ensure a high local signal-to-noise ratio in the compressed images.

Main Methods:

• The proposed algorithm is based on the JPEG compression standard.
• A fuzzy inference system is employed to adapt the normalization of transformed coefficients.
• Local image characteristics of echocardiographic data are analyzed to guide the adaptation process.
• The algorithm was tested on images acquired during actual echocardiography examinations.

Main Results:

• The developed algorithm effectively compresses echocardiographic images.
• The fuzzy inference system successfully adapts the normalization process based on local image features.
• The compression method guarantees a maximum local signal-to-noise ratio, tailored to image sub-blocks.
• The approach preserves diagnostic quality while reducing data size.

Conclusions:

• The proposed algorithm offers an effective method for compressing echocardiographic images.
• Adaptive normalization using fuzzy logic enhances compression efficiency and image quality.
• This technique is suitable for practical echocardiography applications, improving data management.