Related Experiment Videos
A multiprocessor architecture for medical image compression in a PACS environment
J Azpiroz Leehan1, J F Lerallut, I Magaña
1Universidad Autónoma Metropolitana-Iztapalapa, Departamento de Ingeniería Eléctrica, Iztapalapa, D.F., México.
Medical Progress Through Technology
|January 1, 1994
Summary
A new multiprocessor architecture enhances medical image compression using digital signal processors. This system achieves acceptable image quality across various modalities by optimizing compression algorithms for each specific imaging type.
Area of Science:
- Biomedical Engineering
- Computer Science
- Digital Imaging
Background:
- Medical image compression is crucial for efficient storage and transmission.
- Existing compression methods may not optimally handle diverse medical imaging modalities.
Purpose of the Study:
- To introduce a novel multiprocessor architecture for medical image compression.
- To evaluate the performance of this architecture with different compression algorithms on various medical imaging types.
Main Methods:
- Development of a multiprocessor system utilizing digital signal processors on a NuBus-based workstation.
- Implementation and testing of a full-frame cosine transform with bit allocation tables.
- Evaluation across ultrasound and magnetic resonance imaging modalities.
Main Results:
- The proposed architecture successfully applies compression algorithms to medical images.
- Acceptable image quality was achieved when bit allocation tables were tailored to specific imaging modalities.
- Ongoing development includes advanced algorithms like wavelet transforms and vector quantization.
Conclusions:
- The developed multiprocessor architecture offers a flexible platform for medical image compression.
- Algorithm adaptation to specific imaging modalities is key to maintaining high image quality.
- Future work will integrate more advanced, image-characteristic-responsive compression techniques.