Besides transform-based methods mentioned above, non-parametric and parametric model based algorithms [5 7] also achieve good de-noising results.
Scheme 2 gives better performance than the other schemes, especially for ER Model based algorithms.
The choice of the score function defined by Eqs. 6 and 23 has little impact to the ER model based algorithms.
Deformable model based algorithms, such as snakes, converge slowly and often unreliably and also require some form of initialization.
The state-space model based algorithms are widely used in brain machine interfaces to infer the latent state like the kinematics from neural activity.
This algorithm combines the strength of stochastic-model based algorithms and physical models and has been shown to be superior in accuracy and precision to other normalization methods such as MAS, RMA and PerfectMatch [ 62].
Previous studies using Monte Carlo and AXB calculations proved that conventional model based algorithms predicted dose distributions in bone that were closer to D m distributions than to D w distributions [ 34, 42].
In addition, compared with many state-space model based algorithms, such as Kalman filter and point process adaptive filter [ 15, 16, 22, 23], there is no restriction on the posterior distribution of the state, which makes the SMCPP more flexible and suitable for the highly nonlinear systems such as BMIs.
In the rest of this paper, we use scheme 2 for the ER Model based algorithm.
Reconstruction errors were found to be smaller for the equivalent sources model based algorithm and the statistically optimized NAH algorithm.
Reconstructed particle velocity and acoustic pressure fields presented in this study showed that the equivalent sources model based algorithm along with Tikhonov regularization provided the best localization of the sources.
