Research Interest
My main research interests are in the following domains:
- Coded Excitation
- Data Acquisition
- Signal Processing
- Tissue Harmonic Imaging
- Ultrasound Contrast Microbubbles for Imaging and Therapy
- Contrast Enhanced Ultrasound Imaging
- Subharmonic Imaging
- Superharmonic Imaging
- Ultrasonic Characterisation and Imaging of the Trabecular Bone
Tissue Harmonic Imaging
Tissue Harmonic Imaging (THI) technique relies on the second harmonic frequency, which is generated due to the nonlinear distortion of ultrasound waves propagating through biological tissue. There are several methods which are used to extract the second harmonic component from the received echo signal. These techniques can be divided into two types: linear filtering and signal coding. Linear filtering performs well when no spectral overlapping exists amongst the fundamental and second harmonic frequencies but it fails when the signal’s amplitude spectrum overlap.
Similarly signal coding techniques can enhance the magnitude of the required second harmonic and completely remove the unwanted fundamental frequency from the received echo signal but it requires multiple transmissions, causing reduction in system frame rate. Signal coding also fails to cancel out the unwanted fundamental frequency in the presence of motion artifacts, resulting in additional filtering to remove the unwanted frequencies.
This work focuses on signal processing techniques to both encode the excitation waveform to get the better penetration depth and requires only single transmission which helps to avoid frame rate reduction and movement artefact problems.
Similarly signal coding techniques can enhance the magnitude of the required second harmonic and completely remove the unwanted fundamental frequency from the received echo signal but it requires multiple transmissions, causing reduction in system frame rate. Signal coding also fails to cancel out the unwanted fundamental frequency in the presence of motion artifacts, resulting in additional filtering to remove the unwanted frequencies.
This work focuses on signal processing techniques to both encode the excitation waveform to get the better penetration depth and requires only single transmission which helps to avoid frame rate reduction and movement artefact problems.
Pulse Compression of Harmonic Chirp Signals Using The Fractional Fourier Transform
In ultrasound harmonic imaging with chirp-coded excitation, a harmonic matched filter (HMF) is typically used on the received signal to perform pulse compression of the second harmonic component (SHC) to recover signal axial resolution. Designing the HMF for the compression of the SHC is a problematic issue as it requires optimal window selection. In the compressed second harmonic signal, the sidelobe level may increase and the mainlobe width (MLW) widen under a mismatched condition, resulting in loss of axial resolution.
We propose the use of the fractional Fourier transform (FrFT) as an alternative tool to perform compression of the chirp-coded SHC generated as a result of the nonlinear propagation of an ultrasound signal. Experimental results indicate that the FrFT provides a 14% and 23% reduction in the MLW of the compressed signals when compared with harmonic matched and mismatched filtering techniques respectively. The peak and integrated sidelobe levels provided by the FrFT are comparable to the harmonic matched and mismatched filtering techniques. Also the FrFT requires no windowing function or prior knowledge of the initial frequency of the received signal.
We propose the use of the fractional Fourier transform (FrFT) as an alternative tool to perform compression of the chirp-coded SHC generated as a result of the nonlinear propagation of an ultrasound signal. Experimental results indicate that the FrFT provides a 14% and 23% reduction in the MLW of the compressed signals when compared with harmonic matched and mismatched filtering techniques respectively. The peak and integrated sidelobe levels provided by the FrFT are comparable to the harmonic matched and mismatched filtering techniques. Also the FrFT requires no windowing function or prior knowledge of the initial frequency of the received signal.
Overlapped Second Harmonic Chirp Extraction using The Fractional Fourier Transform
In ultrasound harmonic imaging with chirp coded excitation, the axial resolution can be improved by increasing the excitation signal bandwidth. However, increasing the bandwidth will cause overlapping between the received nonlinear second harmonic chirp component (SHCC) and the fundamental component. For the spectrally overlapping harmonics, signal decoding using the second harmonic matched filter (SHMF) typically produces higher range sidelobes level (RSLL), which reduces the image contrast. A multi-pulse detection scheme such as pulse inversion can be used to extract the overlapped SHCC; however it is susceptible to motion artifacts and reduced system frame rate. In this study, the fractional Fourier transform (FrFT) is proposed with chirp coded excitation for the extraction of the overlapped SHCC. The experimental results indicate at least a 13 dB improvement in the RSLL of the FrFT filtered compressed SHCC when compared with the unfiltered compressed SHCC.
Ultrasound Harmonic Imaging using Nonlinear Frequency Modulated Signals
In ultrasound harmonic imaging with linear frequency modulated (LFM) chirp excitation, the sidelobes level in the compressed harmonic chirp signal can be reduced by applying a windowing function.Windowing on the transmitting signal causes reduced penetration depth, whilst windowing on the receiving side results in reduced signal-to-noise ratio (SNR) gain and axial resolution. To optimize the transmitting signal energy and the SNR gain with reduced sidelobes level in the compressed harmonic signal, the use of nonlinear frequency modulated (NLFM) chirp signals are proposed. The NLFM signal and associated second harmonic matched filter are designed using an analytical approach to minimise correlation errors. In all simulations and experiments, the NLFM signal performance is compared with the reference LFM signal of similar sweeping bandwidth and duration. The results indicate at least a 15 dB reduction in the peak sidelobes level of the NFLM compressed second harmonic signal with comparable axial mainlobe width when compared with the LFM compressed harmonic signal.
Investigation of the Subharmonic Response from Contrast Microbubbles using Linear and Nonlinear Frequency Modulated Signals
The use of microbubbles as ultrasound contrast agents (UCA) has been widely investigated in ultrasound contrast imaging. At low acoustic pressure, contrast microbubbles are able to scatter energy not only at the fundamental but also at second harmonic, subharmonic and ultra-harmonic frequencies. These nonlinear harmonic components are extensively exploited in ultrasound contrast imaging to enhance the contrast between the blood and surrounding tissue. Most medical ultrasound imaging systems offer second harmonic imaging to improve the spatial resolution and it is widely used in clinical applications. However, generation of the second harmonic component due to the nonlinear propagation of ultrasound waves through tissue can degrade the contrast of the image.
Ultrasound imaging with the subharmonic component has the potential to improve the contrast-to-tissue ratio (CTR) as it is solely generated by the microbubble contrast agents. In contrast, tissue is not able to generate subharmonic components at typical acoustic pressures used in medical ultrasound imaging. The bandwidth of the subharmonic component is one-half of the fundamental component causing a reduction in the axial resolution. However, the subharmonic component will face less attenuation as it propagates, resulting in an improved penetration depth.
A number of multi-pulse excitation schemes have been proposed at low mechanical index (MI) to take advantage of the nonlinear response of microbubbles and to prevent microbubble destruction. In general, these multi-pulse detection schemes alter the amplitude and/or phase of the excitation signals. These multi-pulse excitation schemes improve the SNR and CTR at low MI, but they are susceptible to motion artifacts and reduce the temporal resolution.
The aim of this study is to experimentally investigate the effect of varying the bandwidth of linear and nonlinear frequency modulated excitation signals and applied acoustic pressure on the subharmonic response from contrast microbubbles.
Ultrasound imaging with the subharmonic component has the potential to improve the contrast-to-tissue ratio (CTR) as it is solely generated by the microbubble contrast agents. In contrast, tissue is not able to generate subharmonic components at typical acoustic pressures used in medical ultrasound imaging. The bandwidth of the subharmonic component is one-half of the fundamental component causing a reduction in the axial resolution. However, the subharmonic component will face less attenuation as it propagates, resulting in an improved penetration depth.
A number of multi-pulse excitation schemes have been proposed at low mechanical index (MI) to take advantage of the nonlinear response of microbubbles and to prevent microbubble destruction. In general, these multi-pulse detection schemes alter the amplitude and/or phase of the excitation signals. These multi-pulse excitation schemes improve the SNR and CTR at low MI, but they are susceptible to motion artifacts and reduce the temporal resolution.
The aim of this study is to experimentally investigate the effect of varying the bandwidth of linear and nonlinear frequency modulated excitation signals and applied acoustic pressure on the subharmonic response from contrast microbubbles.
Superharmonic Imaging
Ultrasound diagnostic imaging based on the nonlinear second harmonic component is now the de-facto standard in clinical practice. A new ultrasound imaging technique called “superharmonic imaging” (SHI) was proposed in the recent years. SHI relies on the third, fourth and fifth harmonic components of the nonlinear received signal. These higher order harmonic components are produced either due to the nonlinear propagation of ultrasound waves through biological tissue at high acoustic pressure or by the nonlinear backscattering from ultrasound contrast agents insonated at their resonance frequency. SHI provides improved axial and lateral resolution with reduced near-field and reverberation artifacts. Also by the incorporation of ultrasound contrast agents, it provides better contrast-to-tissue ratio (CTR) than the standard second harmonic imaging.
The main issues in the SHI are: low signal-to-noise ratio (SNR) of the higher order nonlinear harmonic components; ripple artifacts in SHI originating due to spectral gaps between the harmonic components under narrow bandwidth pulse excitation; and the requirement of large transducer bandwidth and sensitivity to accommodate fundamental to fifth order harmonics of the nonlinear received signal. In this study, linear frequency modulated chirp signals are proposed as an excitation to improve the SNR and axial resolution of the SHI.
The main issues in the SHI are: low signal-to-noise ratio (SNR) of the higher order nonlinear harmonic components; ripple artifacts in SHI originating due to spectral gaps between the harmonic components under narrow bandwidth pulse excitation; and the requirement of large transducer bandwidth and sensitivity to accommodate fundamental to fifth order harmonics of the nonlinear received signal. In this study, linear frequency modulated chirp signals are proposed as an excitation to improve the SNR and axial resolution of the SHI.
Ultrasonic Characterisation and Imaging of the Trabecular Bone Using Coded Excitation
Signal-to-noise ratio (SNR) and penetration depth are the most valuable factors in medical ultrasound imaging and measurement systems. The majority of commercially available medical ultrasound imaging systems utilizes brief pulse excitation, which carries less energy causes low SNR and has limited penetration depth.
Coded signals which are widely used in radar and mobile communication systems have a potential to increase the total excitation energy by increasing the excitation duration whilst preserving the peak power within the acceptable limits, resulting in improved SNR and penetration depth. It is predicted that 15 to 20 dB improvement in SNR can be possible using coded excitation.
Osteoporosis is a skeletal disease characterized by a decrease in bone mass, strength and micro-architectural deterioration of the bone tissue, causing bones to become fragile and vulnerability to fracture. Unlike radiation based techniques, Quantitative ultrasound (QUS) is an emerging alternative diagnostic technique to assess osteoporosis. Quantitative ultrasound has a potential to provide bone micro-architectural information in addition to bone mineral density (BMD). Speed of sound (SOS) and broadband ultrasonic attenuation (BUA) are the important parameters for the assessment of osteoporosis. The aim of this study is it to measure the quantitative ultrasound parameters (SOS and BUA) in trabecular bone (in vitro) using coded excitation.
Coded signals which are widely used in radar and mobile communication systems have a potential to increase the total excitation energy by increasing the excitation duration whilst preserving the peak power within the acceptable limits, resulting in improved SNR and penetration depth. It is predicted that 15 to 20 dB improvement in SNR can be possible using coded excitation.
Osteoporosis is a skeletal disease characterized by a decrease in bone mass, strength and micro-architectural deterioration of the bone tissue, causing bones to become fragile and vulnerability to fracture. Unlike radiation based techniques, Quantitative ultrasound (QUS) is an emerging alternative diagnostic technique to assess osteoporosis. Quantitative ultrasound has a potential to provide bone micro-architectural information in addition to bone mineral density (BMD). Speed of sound (SOS) and broadband ultrasonic attenuation (BUA) are the important parameters for the assessment of osteoporosis. The aim of this study is it to measure the quantitative ultrasound parameters (SOS and BUA) in trabecular bone (in vitro) using coded excitation.