PubMed Health⌕ Search

Biomedical subjects

Sung Shin Kim

Publications and source records attributed to Sung Shin Kim.

7 recordsLinked to original sources

Improvement of spike train decoder under spike detection and classification errors using support vector machine.

The successful decoding of kinematic variables from spike trains of motor cortical neurons is essential for cortical neural prosthesis. Spike trains from each single unit must be extracted from extracellular neural signals and, thus, spike detection and sorting procedure is indispensable but the detection and sorting may involve considerable error. Thus, a decoding algorithm should be robust with respect to spike train errors. Here, we show that spike train decoding algorithms employing nonlinear mapping, especially a support vector machine (SVM), may be more advantageous contrary to previous results which showed that an optimal linear filter is sufficient. The advantage became more conspicuous in the case of erroneous spike trains. Using the SVM, satisfactory training of the decoder could be achieved much more easily, compared to the case of using a multilayer perceptron, which has been employed in previous studies. Tests were performed on simulated spike trains from primary motor cortical neurons with a realistic distribution of preferred direction. The results suggest the possibility that a neuroprosthetic device with a low-quality spike sorting preprocessor can be achieved by adopting a spike train decoder that is robust to spike sorting errors.

Action Potentials↗

Superiority of nonlinear mapping in decoding multiple single-unit neuronal spike trains: a simulation study.

One of the most important building blocks of the brain-machine interface (BMI) based on neuronal spike trains is the decoding algorithm, a computational method for the reconstruction of desired information from spike trains. Previous studies have reported that a simple linear filter is effective for this purpose and that no noteworthy gain is achieved from the use of nonlinear algorithms. In order to test this premise, we designed several decoding algorithms based on the linear filter, and two nonlinear mapping algorithms using multilayer perceptron (MLP) and support vector machine regression (SVR). Their performances were assessed using multiple neuronal spike trains generated by a biophysical neuron model and by a directional tuning model of the primary motor cortex. The performances of the nonlinear algorithms, in general, were superior. The advantages of using nonlinear algorithms were more profound for cases where false-positive/negative errors occurred in spike trains. When the MLPs were trained using trial-and-error, they often showed disappointing performance comparable to that of the linear filter. The nonlinear SVR showed the highest performance, and this may be due to the superiority of SVR in training and generalization.

Action Potentials↗

Comparison of clinical efficacy of Newfactan versus Surfacten for the treatment of respiratory distress syndrome in the newborn infants.

Newfactan is a domestically developed, bovine lung-derived, semi-synthetic surfactant. The aim of this study was to compare the clinical efficacy of Newfactan with that of Surfacten in the treatment of respiratory distress syndrome (RDS). Newfactan or Surfacten was randomly allocated to 492 newborn infants who were diagnosed as RDS and required surfactant instillation in four participating hospitals. The comparisons were made individually in two subsets of infants by birth weight (<1,500 g group [n=253] and > or =1,500 g group [n=239]). Short-term responses to surfactant and acute complications, such as the total doses of surfactant instilled, response type, extubation rate, ventilator settings, changes in respiratory parameters, air leak, patent ductus arteriosus, pulmonary hemorrhage, and intraventricular hemorrhage, and mortality during the 96 hr after surfactant instillation were measured. Long-term outcome and complications, such as total duration of intubation, bronchopulmonary dysplasia and periventricular leukomalacia, and ultimate mortality were measured. There were no significant differences in demographic and perinatal variables, shortterm responses to surfactant and acute complications, and long-term outcome and complications between Newfactan and Surfacten in both birth weight groups. We concluded that Newfactan was comparable to Surfacten in the clinical efficacy in the treatment of RDS in both birth weight groups.

Female↗

Decreasing incidence of chronic lung disease despite the gradual reduction of postnatal dexamethasone use in very low birth weight infants.

Dexamethasone has been widely used in very low birth weight infants (VLBWI) weighing less than 1,500 g at birth for the prevention or treatment of chronic lung disease (CLD). Recently, however the use of dexamethasone is being reduced, as its association with abnormal neurodevelopmental outcome is known. On the other hand, there have been persistent concerns about the increased risk of CLD according to the reduction of postnatal dexamethasone use. Hence, we did a retrospective cohort study to delineate the change in the incidence of CLD according to the reduction of dexamethasone use in VLBWI. The medical records of 559 VLBWI admitted to neonatal intensive care unit at Samsung Medical Center between November 1994 and December 2002 were reviewed with a focus on the use of postnatal dexamethasone and the incidence of CLD. The use of postnatal dexamethasone has significantly decreased over the study period. Especially, the use of high-dose regimen has markedly decreased. The day when postnatal dexamethasone therapy was begun has also been significantly delayed. The incidence of CLD has significantly decreased over the same period. In conclusion, the incidence of CLD has not increased despite the decreased use of postnatal dexamethasone.

Chronic Disease↗

Effects of MK-801 (dizocilpine) on brain cell membrane function and energy metabolism in experimental Escherichia coli meningitis in the newborn piglet.

We evaluated the efficacy of non-competitive N-methyl-D-aspartate receptor antagonist MK-801 (dizocilpine) as an adjuvant therapy in experimental neonal bacterial meningitis. Meningitis was induced by injecting 10(6) colony forming units of Escherichia coli into the cisterna magna. MK-801 3 mg/kg was given as a bolus intravenous injection, 30 min before the induction of meningitis. MK-801 did not down-modulate the inflammatory parameters, such as increased intracranial pressure, cerebrospinal fluid (CSF) leukocytosis, increased lactate and TNF-alpha levels in the CSF, and hypoglycorrhachia observed in the meningitis group. MK-801 did not significantly attenuate the elevated glutamate concentration in the CSF. However, MK-801 showed some neuroprotective effects as evidenced by significant attenuation of cerebral lipid peroxidation products (conjugated dienes) and increase of brain high-energy phosphate compounds (ATP and PCr). Improvement in cerebral cortical cell membrane Na+, K+ -ATPase activity did not reach a statistical significance. These results suggest that MK-801 was effective in ameliorating brain injury in neonatal bacterial meningitis, although it failed to attenuate the inflammatory responses.

Animals↗

Detrimental effects of N(omega) nitro-L-arginine methyl ester (L-NAME)in experimental Escherichia coli sepsis in the newborn piglet.

The role of nitric oxide during neonatal sepsis is complex. We tested the hypothesis that nonselective inhibition of nitric oxide synthase with N(omega) -nitro-L-arginine methyl ester (L-NAME) is detrimental during the early phase of experimental sepsis in the newborn piglet. Newborn piglets were divided into four groups: 6 in the control group, 6 in the L-NAME control group, 12 in the sepsis group (SG), and 11 in the sepsis with L-NAME group (NS). Sepsis was induced by intravenous injection of 10(8) colony forming units of Escherichia coli. L-NAME 10 mg/kg was given intravenously 60 min before the induction of sepsis. The survival rate of piglets after 4 hr was 27% in NS, while it was 100% in other groups. Systemic hypotension, observed in both SG and NS, were more profound in NS. Leukopenia was observed in both SG and NS. Thrombocytopenia, prolongation of prothrombin time and activated partial thromboplastin time, and increase in thrombin-antithrombin complexes were observed only in NS. Decreased PaO2 /FiO2 ratio, arterial pH and base excess, and increased blood lactate levels observed in both SG and NS, but were more profound in NS. These findings suggest that nonselective inhibition of nitric oxide synthase with L-NAME is detrimental during the early phase of experimental neonatal sepsis.

Animals↗

Effects of dopamine infusion on cerebral blood flow, brain cell membrane function and energy metabolism in experimental Escherichia coli meningitis in the newborn piglet.

In the present study, we tested whether maintenance of adequate cerebral perfusion pressure (CPP) by pharmacologically preventing systemic hypotension with dopamine infusion would prevent cerebral ischemia and attenuate energy depletion and neuronal injury even though intracranial pressure remains elevated in a newborn piglet meningitis model. Cerebral blood flow, measured at the end of the experiment using fluorescent microspheres, was significantly increased by dopamine infusion. The decreased cerebral cortical cell membrane Na+, K+ -ATPase activity and increased lipid peroxidation products, indicative of meningitis-induced brain damage, were significantly attenuated by dopamine infusion. Dopamine also significantly attenuated the meningitis-induced reduction in both brain ATP and phosphocreatine levels and the increase in brain lactate level. In summary, maintenance of adequate CPP with dopamine prevented cerebral ischemia, reduced cerebral energy depletion, and attenuated brain injury in neonatal bacterial meningitis.

Animals↗