Fast, limited-data photoacoustic imaging for multiplexed systems using a frequency-domain estimation technique.

摘要

PURPOSE: A new frequency-domain estimation algorithm has been developed that uses a priori information to simultaneously improve imaging quality and time resolution in photoacoustic tomography with incomplete data sets. METHODS: The method involves application of a single-stage Wiener optimal filter to augment data sets by interpolation between measurement locations using relationships determined in a reference scan. The filter can be applied in real-time using FFT methods using either fixed or dynamic references and used with any imaging algorithm. The performance of the method is compared to a modified version of constrained backprojection algorithms using simulations and experimental investigations. RESULTS: Simulations demonstrate the effectiveness of the approach for tracking dynamic photoacoustic activity for data sets with limited views (90 degrees) or tomographic views with a reduced number of acquisition angles at any given time (< or = 32). Experimental data of contrast uptake and washout using a 512-element curved transducer with 8:1 electronic multiplexing with the algorithm demonstrate full two-dimensional tomographic imaging with a temporal resolution better than 130 ms. CONCLUSIONS: The estimation algorithm enables high spatial resolution, real-time imaging of dynamic physiological events or volumetric regions for photoacoustic systems employing multiplexing or scanning.