Ultrasound color-flow imaging on a programmable system.

Area of Science:

• Medical Imaging
• Ultrasound Technology
• Signal Processing

Background:

• Color-flow imaging is crucial for real-time blood flow visualization but is computationally intensive.
• Conventional ultrasound systems rely on fixed-function hardware to manage computational demands.
• Programmable platforms offer potential for greater flexibility and efficiency in ultrasound processing.

Purpose of the Study:

• To present a novel ultrasound system where color-flow processing is entirely supported on a programmable platform.
• To evaluate the performance and clinical feasibility of a software-based color-flow imaging system.
• To demonstrate the advantages of programmability in ultrasound system design.

Main Methods:

• Development of a color-flow processing system implemented primarily in C language (approx. 95% of modules).
• Utilized a single processor for initial performance evaluation.
• Input data: 192 x 512 x 8 samples for color flow, 384 x 512 for B mode; output image size: 600 x 420.

Main Results:

• Achieved a frame rate of 7.9 frames/s on a single processor with specified input/output parameters.
• Demonstrated that the programmable system can provide equivalent clinical functionality to conventional ultrasound systems.
• Indicated that additional processors can enhance performance (higher frame rates or more data).

Conclusions:

• A programmable ultrasound system can successfully perform clinical color-flow imaging.
• The proposed system offers superior flexibility and efficiency compared to traditional fixed-function hardware systems.
• Software-based implementation on programmable platforms is a viable and advantageous approach for advanced ultrasound functionalities.