Wavelet domain frequency interpolation for photo-acoustic tomography

Photo-acoustic imaging has grown tremendously in the past few years, both in instrumentation and application. Due to its higher penetration, resolution and contrast it is gaining importance as a promising imaging modality. Commercially available transducers are typically bandlimited and thereby limit the frequency range of the received ultrasound signal. One approach is to use multiple bandwidth transducers but doing so increases the cost of the system. A wavelet based frequency interpolation method is proposed in this work. Low frequency band limited transducer is used in the experiments, and the missing high frequency components are estimated using an interpolation algorithm in the wavelet domain. A database of synthetic wideband photoacoustic data is constructed first. This database is then used to estimate the most likely high frequency components of the narrowband test signal. The approach provides better resolution at a low cost. It is shown that the reconstructed image has better resolution than multiple bandwidth transducer approach. This reduces the system complexity and along with compressive sensing techniques a hand held clinical photo-acoustic probe can be built.