Estimation of side lobes in ultrasound imaging systems

Abstract

Purpose

We propose a new method of estimating the side lobe levels from received ultrasound channel data.

Methods

Ultrasound signals impinging on an array transducer manifest themselves as sinusoids whose spatial frequency varies as a function of the incident angle. The received channel data from the main lobe direction have a spatial frequency of zero when considered across all the channels because individual channel data have the same phase in the temporal domain, while those from the side lobe directions have a spatial frequency that is not zero and varies with the ultrasound beam incident direction. Thus, the received channel data can be modeled as a sinusoidal waveform that has a specific frequency depending on the incident angle. The side lobe components are modeled as a sum of sinusoidal waves that have a frequency equal to an integer plus a half. We estimated the side lobe waveform that has an integer-plus-a-half frequency by adaptively varying the window length in the spectrum estimation. The effect of side lobes on ultrasound image can be reduced by subtracting the estimated side lobe waveform in receive focusing.

Results

To confirm the efficacy of the proposed method, ultrasound field simulations and experiments were carried out. We observed that the 1st to 5th spatial frequency side lobes estimated using the proposed method were present at the same positions as the side lobes in the field response. By subtracting the waveforms of the 1st to 20th spatial frequency side lobes, we reduced the side lobe levels by up to 14 dB. Experiments on wire phantoms in a water tank also confirmed that the proposed method can estimate and reduce the side lobe levels.

Conclusions

We have proposed a new method of estimating and canceling side lobes that is validated through both simulation and experiment.