Search

There has been considerable research interest in reconstructing three-dimensional structure of biological cells using microscope. Off-axis illumination system is generally used for low or middle magnification, but many numbers of projections are usually required for good image reconstruction. This paper proposes a new algorithm such that three-dimensional sample can be reconstructed using only a few projection if the cells are distributed sparsely along phantom. [ABSTRACT FROM AUTHOR]

For high quality MR imaging of time-varying objects such as beating heart or brain hemodynamics, fast signal acquisition is required without sacrificing the spatial resolution. Recently, we proposed a novel parallel imaging algorithm for dynamically varying objects. The new algorithm, called x-f SENSE (x-f space SENSitivity Encoding), optimally reduces the data acquisition rate by exploiting the temporal redundancies and sensitivity diversities in a phase array coil. The acceleration factor of the x-f SENSE is up to the product of those of the parallel imaging algorithm and the optimal spatio-temporal encoding algorithm. Furthermore, the reconstruction algorithm of x-f SENSE is as simple as the ordinary SENSE. Due to its simplicity and the optimal acceleration factor, x-f SENSE can be immediately used in practice. However, reconstruction image quality is very dependent on the measurement noise level. This paper is devoted to evaluate the reconstruction noise variance of the x-f SENSE by analyzing the coil sensitivity and noise covariance between the coils. Based on the performance in noisy environments, we again confirm that the x-f SENSE is a very powerful algorithm for fast varying dynamic images in practice. [ABSTRACT FROM AUTHOR]