研究目的
To develop a better compression algorithm for medical images that yields good compression ratio while preserving image quality, especially in diagnostically important regions.
研究成果
The proposed methodology provides an efficient presentation of images while preserving 2D singularities along the curves through multi-resolution analysis. It achieves an appreciable compression ratio compared to conventional transforms and preserves diagnostically important information. Future work could explore different transformations and filters to enhance edge preservation in smooth areas.
研究不足
The proposed method may not preserve edges of smooth areas (low-level components) as effectively, and the levels for multi-resolution analysis are not dynamically adjusted based on the intensity of edges in the image.
1：Experimental Design and Method Selection:
The proposed method uses Ripplet transform for preserving inconsistencies along curves and Huffman encoder for compressing the resulting coefficients.
2：Sample Selection and Data Sources:
Digital Imaging and Communication in Medicine (DICOM) datasets from various diagnostic centers including CT, MRI, X-ray, USG, PET, and PET-CT datasets.
3：List of Experimental Equipment and Materials:
Not explicitly mentioned.
4：Experimental Procedures and Operational Workflow:
The algorithm involves decomposing medical images into low- and high-frequency components using 9/7 wavelet filter, normalizing the components, applying ripplet transform, and then Huffman encoding.
5：Data Analysis Methods:
Performance is assessed using Mean Square Error (MSE), Peak Signal-to-Noise Ratio (PSNR), Compression Ratio (CR), and Response Time.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
