Lennox-Gastaut syndrome (LGS) is a devastating and refractory generalized epilepsy affecting children and adolescents. In this study we report the results of resective surgery in 18 patients with LGS phenotype who underwent single-lobe/lesionectomy or multilobe resection plus multiple subpial transection and/or callosotomy. After surgery, seven patients became completely seizure-free (Engel Class I) and five almost seizure-free (Engel Class II). Additional four had significant seizure control (Engel Class III), and two had no change in seizure frequency (Engel Class IV). Of the 4 patients without any lesion on brain MRI, 2 ended with Engel Class II, 1 with III and the other with IV in Engels' classification. Mean intelligence quotient (IQ) increased from 56.1 ± 8.1 (mean ± SD) before operation to 67.4 ± 8.2 (mean ± SD) after operation, a significant improvement (P = 0.001). Results also indicated that the younger the patient at surgery, or the shorter the interval between onset of seizure and resective operation, the better the intellectual outcome. Our data suggest that resective epilepsy surgery can be successful in patients with LGS phenotype as long as the EEG shows dominance of discharges in one hemisphere and corresponding ipsilateral imaging findings, even with contralateral ictal discharges.

No abstract

In this paper, we propose a new weak maneuvering target detection method in sea environment with Radonfractional Fourier transform (RFrFT) canceller, which can efficiently suppress the sea clutter and detect targets with low computations. To accomplish the sea clutter suppression, we design a canceller based on RFrFT in the detection process, which utilizes different representations of maneuvering target and sea clutter. A non-uniform searching strategy with nonuniform searching interval of fractional order is presented to reduce the computation of the detection process. Moreover, to make the RFrFT canceller also effective for multiple target case, we propose a band-stop design on the basis of the RFrFT convolution theorem by making use of the detection results after using the canceller. After that, the detailed detection algorithm is described. Experimental results on simulated and real data show that the proposed method is superior compared with other methods.

No abstract

No abstract