A Sub-Nyquist, Variable Sampling, High-Frequency Phased Array Beamformer.

Area of Science:

• Ultrasound imaging
• Digital signal processing
• Medical instrumentation

Background:

• Traditional ultrasound systems require high sampling rates, increasing data processing demands.
• Variable sampling techniques have been explored but present limitations in efficiency and accuracy.
• Developing efficient digital beamformers is crucial for advancing ultrasound imaging capabilities.

Purpose of the Study:

• To describe a digital receive beamformer utilizing a "one sample per pixel" variable sampling technique.
• To evaluate the performance of this new sampling method in reducing data rates and maintaining accuracy.
• To demonstrate the feasibility and effectiveness of the beamformer in real-world ultrasound imaging applications.

Main Methods:

• Implementation of a "one sample per pixel" variable sampling technique in a digital receive beamformer.
• Field-programmable gate array (FPGA) implementation of the beamforming algorithm.
• Testing and verification using a 45-MHz 64-element phased array, imaging wire targets, and in vivo human wrist scans.

Main Results:

• The "one sample per pixel" method reduced sampling rates by a factor of 3 and data capture rates by a factor of 2.
• Accurate estimation of broadband pulse envelopes for bandwidths up to 83.0% was achieved.
• Maximum transmit and receive delay errors were measured below ±1.0 ns, with images showing ~55 dB dynamic range.

Conclusions:

• The described digital beamformer with "one sample per pixel" variable sampling offers significant data reduction.
• This technique enables accurate broadband pulse envelope estimation, crucial for high-resolution ultrasound imaging.
• The successful implementation and testing validate its potential for advanced ultrasound diagnostic systems.