Spinal cord stimulation (SCS) currently relies on extradural electrode arrays that are separated from the spinal cord surface by a highly conducting layer of cerebrospinal fluid. It has recently been suggested that intradural placement of the electrodes in direct contact with the pial surface could greatly enhance the specificity and efficiency of stimulation. The present computational study aims at quantifying and comparing the electrical current distributions as well as the spatial recruitment profiles resulting from extraand intra-dural electrode arrangements. The electrical potential distribution is calculated using a 3D finite element model of the human thoracic spinal canal. The likely recruitment areas are then obtained by using the potential as input to an equivalent circuit model of the pre-threshold axonal response. The results show that the current threshold to recruitment of axons in the dorsal column is more than an order of magnitude smaller for intradural than extradural stimulation. Intradural placement of the electrodes also leads to much higher contrast between the stimulation thresholds for the dorsal root entry zone and the dorsal column, allowing better focusing of the stimulus.

No abstract

The proper selection of parameters, kernel parameter g, penalty factor c, non-sensitive coefficient p of Support Vector Regression (SVR) model can optimize SVR's performance. The most commonly used approach is grid search. However, when the data set is large, a terribly long time will be introduced. Thus, we propose an improved grid algorithm to reduce searching time by reduce the number of doing cross-validation test. Firstly, the penalty factor c could be calculated by an empirical formula. Then the best kernel parameter g could be found by general grid search algorithm with the achieved c and a p-value selected randomly within a range. According to the achieved c and p, the grid search algorithm is used again to search the best non-sensitive coefficient p. Experiments on 5 benchmark datasets illustrate that the improved algorithm can reduce training time markedly in a good prediction accuracy.

No abstract

No abstract