Kernel machines for epilepsy diagnosis via EEG signal classification: A comparative study

Objective: We carry out a systematic assessment on a suite of kernel-based learning machines while coping with the task of epilepsy diagnosis through automatic electroencephalogram (EEG) signal classification. Methods and materials: The kernel machines investigated include the standard support vector machine (SVM), the least squares SVM, the Lagrangian SVM, the smooth SVM, the proximal SVM, and the relevance vector machine. An extensive series of experiments was conducted on publicly available data, whose clinical EEG recordings were obtained from five normal subjects and five epileptic patients. The performance levels delivered by the different kernel machines are contrasted in terms of the criteria of predictive accuracy, sensitivity to the kernel function/parameter value, and sensitivity to the type of features extracted from the signal. For this purpose, 26 values for the kernel parameter (radius) of two well-known kernel functions (namely. Gaussian and exponential radial basis functions) were considered as well as 21 types of features extracted from the EEG signal, including statistical values derived from the discrete wavelet transform, Lyapunov exponents, and combinations thereof. Results: We first quantitatively assess the impact of the choice of the wavelet basis on the quality of the features extracted. Four wavelet basis functions were considered in this study. Then, we provide the average accuracy (i.e., cross-validation error) values delivered by 252 kernel machine configurations; in particular, 40%/35% of the best-calibrated models of the standard and least squares SVMs reached 100% accuracy rate for the two kernel functions considered. Moreover, we show the sensitivity profiles exhibited by a large sample of the configurations whereby one can visually inspect their levels of sensitiveness to the type of feature and to the kernel function/parameter value. Conclusions: Overall, the results evidence that all kernel machines are competitive in terms of accuracy, with the standard and least squares SVMs prevailing more consistently. Moreover, the choice of the kernel function and parameter value as well as the choice of the feature extractor are critical decisions to be taken, albeit the choice of the wavelet family seems not to be so relevant. Also, the statistical values calculated over the Lyapunov exponents were good sources of signal representation, but not as informative as their wavelet counterparts. Finally, a typical sensitivity profile has emerged among all types of machines, involving some regions of stability separated by zones of sharp variation, with some kernel parameter values clearly associated with better accuracy rates (zones of optimality). (C) 2011 Elsevier B.V. All rights reserved.

Severe Lactic Acidosis Treated With Prolonged Hemodiafiltration in a Disseminated Histoplasmosis

Infections with Histoplasma are rarely seen in immunocompromized patients. We report the case of a renal transplant recipient who presented with disseminated histoplasmosis 3.5 years after transplant He presented severe lactic acidosis (LA), sepsis complicated by circulatory failure, renal failure, and liver dysfunction. We describe the successful use of continuous venovenous hemodiafiltration (CVVHDF) with regional citrate anticoagulation, treatment that stabilized our patient until infectious focus was identified and treated. The lactate was decreasing, concomitant with hemodynamic improvement, with reduction and suspension of the norepinephrine. The serum lactate level normalized 52 hours after CVVHDF initiated (from 28.9 to 2.2 mmol/L). Continuous renal replacement therapy was safely applied and can be recommended as an efficient method on adjuvant treatment of hyperlactatemia. ASAIO Journal 2009; 55:123-125.

Support vector machines for detecting age-related changes in running kinematics

Age-related changes in running kinematics have been reported in the literature using classical inferential statistics. However, this approach has been hampered by the increased number of biomechanical gait variables reported and subsequently the lack of differences presented in these studies. Data mining techniques have been applied in recent biomedical studies to solve this problem using a more general approach. In the present work, we re-analyzed lower extremity running kinematic data of 17 young and 17 elderly male runners using the Support Vector Machine (SVM) classification approach. In total, 31 kinematic variables were extracted to train the classification algorithm and test the generalized performance. The results revealed different accuracy rates across three different kernel methods adopted in the classifier, with the linear kernel performing the best. A subsequent forward feature selection algorithm demonstrated that with only six features, the linear kernel SVM achieved 100% classification performance rate, showing that these features provided powerful combined information to distinguish age groups. The results of the present work demonstrate potential in applying this approach to improve knowledge about the age-related differences in running gait biomechanics and encourages the use of the SVM in other clinical contexts. (C) 2010 Elsevier Ltd. All rights reserved.

Evaluation of alternative technical markers for the pelvic coordinate system

In this study, we evaluated alternative technical markers for the motion analysis of the pelvic segment. Thirteen subjects walked eight times while tri-dimensional kinematics were recorded for one stride of each trial. Five marker sets were evaluated, and we compared the tilt, obliquity, and rotation angles of the pelvis segment: (1) standard: markers at the anterior and posterior superior iliac spines (ASIS and PSIS); (2) markers at the PSIS and at the hip joint centers, HJCs (estimated by a functional method and described with clusters of markers at the thighs); (3) markers at the PSIS and HJCs (estimated by a predictive method and described with clusters of markers at the thighs); (4) markers at the PSIS and HJCs (estimated by a predictive method and described with skin-mounted markers at the thighs based on the Helen-Hayes marker set); (5) markers at the PSIS and at the iliac spines. Concerning the pelvic angles, evaluation of the alternative technical marker sets evinced that all marker sets demonstrated similar precision across trials (about 1 degrees) but different accuracies (ranging from 1 degrees to 3 degrees) in comparison to the standard marker set. We suggest that all the investigated marker sets are reliable alternatives to the standard pelvic marker set. (C) 2009 Elsevier Ltd. All rights reserved.

Joint forces and torques when walking in shallow water

This study reports for the first time an estimation of the internal net joint forces and torques on adults` lower limbs and pelvis when walking in shallow water, taking into account the drag forces generated by the movement of their bodies in the water and the equivalent data when they walk on land. A force plate and a video camera were used to perform a two-dimensional gait analysis at the sagittal plane of 10 healthy young adults walking at comfortable speeds on land and in water at a chest-high level. We estimated the drag force on each body segment and the joint forces and torques at the ankle, knee, and hip of the right side of their bodies using inverse dynamics. The observed subjects` apparent weight in water was about 35% of their weight on land and they were about 2.7 times slower when walking in water. When the subjects walked in water compared with walking on land, there were no differences in the angular displacements but there was a significant reduction in the joint torques which was related to the water`s depth. The greatest reduction was observed for the ankle and then the knee and no reduction was observed for the hip. All joint powers were significantly reduced in water. The compressive and shear joint forces were on average about three times lower during walking in water than on land. These quantitative results substantiate the use of water as a safe environment for practicing low-impact exercises, particularly walking. (C) 2011 Elsevier Ltd. All rights reserved.

Peak torque and rate of torque development in elderly with and without fall history

Background Falls are one of the greatest concerns among the elderly A number of studies have described peak torque as one of the best fall-related predictor. No studies have comprehensively focused on the rate of torque development of the lower limb muscles among elderly fallers. Then, the aim of this study was to determine the relationship between muscle peak torque and rate of torque development of the lower limb joints in elderly with and without fall history It was also aimed to determine whether these parameters of muscle performance (i e, peak torque and rate of torque development) are related to the number of falls. Methods: Thirty-one women volunteered to participate in the study and were assigned in one of the groups according to the number of falls over the 12 months that preceded the present Then, participants with no fall history (Cl; n = 13; 67.6[7.5] years-old), one fall (GII; n = 8, 66 0[4 91 years-old) and two or more falls (GIII, n = 10; 67.8[8.8] years-old) performed a number of lower limb maximal isometric voluntary contractions from which peak torque and rate of torque development were quantified Findings. Primary outcomes indicated no peak torque differences between experimental groups in any lower limb joint. The rate of torque development of the knee flexor muscles observed in the non-fallers (Cl) was greater than that observed in the fallers (P < 0.05) and had a significant relationship with the number of falls (P < 0 05) Interpretation. The greater knee flexor muscles` rate of torque development found in the non-fallers in comparison to the fallers indicated that the ability of the elderly to rapidly reorganise the arrangement of the lower limb may play a significant role in allowing the elderly to recover balance after a trip. Thus, training stimulus aimed to improve the rate of torque development may be more beneficial to prevent falls among the elderly than other training stimulus, which are not specifically designed to improve the ability to rapidly produce large amounts of torque (C) 2010 Published by Elsevier Ltd

Strength training improves fall-related gait kinematics in the elderly: a randomized controlled trial

Background: Falls are one of the greatest concerns among the elderly. Among a number of strategies proposed to reduce the risk of falls, improving muscle strength has been applied as a successful preventive strategy. Although it has been suggested as a relevant strategy, no studies have analyzed how muscle strength improvements affect the gait pattern. The aim of this study was to determine the effects of a lower limb strength training program on gait kinematics parameters associated with the risk of falls in elderly women. Methods: Twenty seven elderly women were assigned in a balance and randomized order into an experimental (n = 14: age = 61.1 (4.3) years, BMI = 26.4 (2.8) kg m(-2)) and a control (n = 13; age = 61.6 (6.6) years; BMI = 25.9 (3.0) kg m(-2)) group. The EG performed lower limb strength training during 12 weeks (3 days per week), being training load increased weekly. Findings: Primary outcomes were gait kinematics parameters and maximum voluntary isometric contractions at pre- and post-training period. Secondary outcomes were training load improvement weekly and one repetition maximum every two weeks. The I maximal repetition increment ranged from 32% to 97% and was the best predictor of changes in gait parameters (spatial, temporal and angular variables) after training for the experimental group. Z-score analysis revealed that the strength training was effective in reversing age-related changes in gait speed, stride length, cadence and toe clearance, approaching the elderly to reference values for healthy young women. Interpretation: Lower limb strength training improves fall-related gait kinematic parameters. Thus, strength training programs should be recommended to the elderly women in order to change their gait pattern towards young adults. (C) 2009 Elsevier Ltd. All rights reserved.

Effect of eight weeks of strength training on fatigue resistance in men and women

The purpose of the present study was to compare the effects of eight weeks of strength training on fatigue resistance in men and women. Thirty-three men and twenty-three women performed eight weeks of strength training in three weekly sessions. Subjects performed four sets using 80% of 1-RM tests on bench press, squat and arm curl. Fatigue index (FI) was used for analysis of decline in motor performance along the sets. The sum of the number of repetitions accomplished in the four sets in each exercise was used to indicate the fatigue resistance. Anova or Ancova two-way ( time x gender) was employed for statistical analysis ( P < 0.05). Eight weeks of strength training increased significantly 1-RM strength, fatigue resistance and total number of repetitions in both genders. FI decreased significantly in both genders after training ( men = 50% vs. women = Time x gender interaction was observed in the total number of repetitions in squat ( P = 0.04) and arm curl exercises, regarding gains to women ( P = 0.01). In conclusion, eight weeks of ST improved strength, FR, FI and total number of repetitions performed. However, women obtained greater adaptations than men.

Profiles of xylose reductase, xylitol dehydrogenase and xylitol production under different oxygen transfer volumetric coefficient values

BACKGROUND: Xylitol is a sugar alcohol (polyalcohol) with many interesting properties for pharmaceutical and food products. It is currently produced by a chemical process, which has some disadvantages such as high energy requirement. Therefore microbiological production of xylitol has been studied as an alternative, but its viability is dependent on optimisation of the fermentation variables. Among these, aeration is fundamental, because xylitol is produced only under adequate oxygen availability. In most experiments with xylitol-producing yeasts, low oxygen transfer volumetric coefficient (K(L)a) values are used to maintain microaerobic conditions. However, in the present study the use of relatively high K(L)a values resulted in high xylitol production. The effect of aeration was also evaluated via the profiles of xylose reductase (XR) and xylitol clehydrogenase (XD) activities during the experiments. RESULTS: The highest XR specific activity (1.45 +/- 0.21 U mg(protein)(-1)) was achieved during the experiment with the lowest K(L)a value (12 h(-1)), while the highest XD specific activity (0.19 +/- 0.03 U mg(protein)(-1)) was observed with a K(L)a value of 25 h(-1). Xylitol production was enhanced when K(L)a was increased from 12 to 50 h(-1), which resulted in the best condition observed, corresponding to a xylitol volumetric productivity of 1.50 +/- 0.08 g(xylitol) L(-1) h(-1) and an efficiency of 71 +/- 6.0%. CONCLUSION: The results showed that the enzyme activities during xylitol bioproduction depend greatly on the initial KLa value (oxygen availability). This finding supplies important information for further studies in molecular biology and genetic engineering aimed at improving xylitol bioproduction. (C) 2008 Society of Chemical Industry

Batch cooling crystallization of xylitol produced by biotechnological route

BACKGROUND: This work deals with the xylitol production by biotechnological routes emphasizing the purification process using crystallization. RESULTS: Xylitol volumetric productivity of 0.665 g L(-1) h(-1) and yield of 0.7024 g g(-1) were obtained after 92 h fermentation. The fermented broth (61.3 g L(-1) xylitol) was centrifuged, treated and concentrated obtain a syrup (745.3 g L(-1) xylitol) which was crystallized twice, xylitol crystals with 98.5-99.2% purity being obtained. CONCLUSION: The hypothetical distribution obtained permits the determination of modeling parameters, which make possible the estimation of crystal dominant size from different initial experimental conditions. (C) 2008 Society of Chemical Industry

Clarification of a wheat straw-derived medium with ion-exchange resins for xylitol crystallization

BACKGROUND: Xylitol bioproduction from lignocellulosic residues comprises hydrolysis of the hemicellulose, detoxification of the hydrolysate, bioconversion of the xylose, and recovery of xylitol from the fermented hydrolysate. There are relatively few reports on xylitol recovery from fermented media. In the present study, ion-exchange resins were used to clarify a fermented wheat straw hemicellulosic hydrolysate, which was then vacuum-concentrated and submitted to cooling in the presence of ethanol for xylitol crystallization. RESULTS: Sequential adsorption into two anion-exchange resins (A-860S and A-500PS) promoted considerable reductions in the content of soluble by-products (up to 97.5%) and in medium coloration (99.5%). Vacuum concentration led to a dark-colored viscous solution that inhibited xylitol crystallization. This inhibition could be overcome by mixing the concentrated medium with a commercial xylitol solution. Such a strategy led to xylitol crystals with up to 95.9% purity. The crystallization yield (43.5%) was close to that observed when using commercial xylitol solution (51.4%). CONCLUSION: The experimental data demonstrate the feasibility of using ion-exchange resins followed by cooling in the presence of ethanol as a strategy to promote the fast recovery and purification of xylitol from hemicellulose-derived fermentation media. (c) 2008 Society of Chemical Industry.

Production, microstructural characterization and mechanical properties of as-cast Ti-10Mo-xNb alloys

Beta titanium (Ti) alloys are one of the most promising groups of Ti alloys for biomedical applications. This work presents the production, microstructural characterization, and mechanical properties of as-cast Ti-10Mo-xNb (x = 0, 3, 6, 9, 20, and 30) alloys. They were produced via arc melting and characterized by scanning electron microscopy and X-ray diffraction. The density of each alloy was measured by the Archimedes method and the mechanical characterization was carried out by using the Vickers microhardness test and Young`s modulus measurements. The results show a near beta microstructure for niobium (Nb) contents lower or equal to 9 wt% while beta single-phase microstructure was obtained for higher Nb additions. The microhardness values decreased with the increase of Nb content in the alloys. The elastic modulus values of Ti-10Mo-3Nb and Ti-10Mo-20Nb alloys were lower than those of cp Ti and Ti-6Al-4V.

Evaluation of hydrodynamic parameters of a fluidized-bed reactor with immobilized yeast

The fluidized bed reactor has successfully been used to perform biotechnological processes addressed to the production of high added value. The present work evaluates hydrodynamic parameters of a bench-scale fluidized bed reactor with cells of the yeast Candida guilliermondii immobilized either in calcium alginate beads or in polyvinyl alcohol (PVA). The effects of the following variables on cell immobilization were evaluated at 30 degrees C and feeding a synthetic medium containing 50 g L-1 xylose: total particle density (cells plus support), terminal velocity, particle drag force, minimum fluidization velocity and bed porosity. According to the results obtained, the reactor was shown to operate like a fixed-bed bioreactor at xi < 0.5 and a fluidized bed bioreactor at xi > 0.5. The maximum flow rate needed to obtain maximum bed fluidization in the reactor was equal to the terminal velocity of the immobilized cell particles. Particles of cells immobilized within these supports showed values of drag coefficient lower than those reported for other high-density supports. The evaluation of these hydrodynamic characteristics lead to an adequate bed fluidization inside the reactor, thus improving oxygen transference and availability in the fermentation medium, making the process more viable for future scale-up. (c) 2008 Society of Chemical Industry.

Adsorption kinetic, thermodynamic and desorption studies of phosphate onto hydrous niobium oxide prepared by reverse microemulsion method

A type of Nb(2)O(5)center dot 3H(2)O was synthesized and its phosphate removal potential was investigated in this study. The kinetic study, adsorption isotherm, pH effect, thermodynamic study and desorption were examined in batch experiments. The kinetic process was described by a pseudo-second-order rate model very well. The phosphate adsorption tended to increase with a decrease of pH. The adsorption data fitted well to the Langmuir model with which the maximum P adsorption capacity was estimated to be 18.36 mg-Pg(-1). The peak appearing at 1050 cm(-1) in IR spectra after adsorption was attributed to the bending vibration of adsorbed phosphate. The positive values of both Delta H degrees and Delta S degrees suggest an endothermic reaction and increase in randomness at the solid-liquid interface during the adsorption. Delta G degrees values obtained were negative indicating a spontaneous adsorption process. A phosphate desorbability of approximately 68% was observed with water at pH 12, which indicated a relatively strong bonding between the adsorbed phosphate and the sorptive sites on the surface of the adsorbent. The immobilization of phosphate probably occurs by the mechanisms of ion exchange and physicochemical attraction. Due to its high adsorption capacity, this type of hydrous niobium oxide has the potential for application to control phosphorus pollution.

Improved xylitol production in media containing phenolic aldehydes: application of response surface methodology for optimization and modeling of bioprocess

BACKGROUND: The combined effects of vanillin and syringaldehyde on xylitol production by Candida guilliermondii using response surface methodology (RSM) have been studied. A 2(2) full-factorial central composite design was employed for experimental design and analysis of the results. RESULTS: Maximum xylitol productivities (Q(p) = 0.74 g L(-1) h(-1)) and yields (Y(P/S) = 0.81 g g(-1)) can be attained by adding only vanillin at 2.0 g L(-1) to the fermentation medium. These data were closely correlated with the experimental results obtained (0.69 +/- 0.04 g L(-1) h(-1) and 0.77 +/- 0.01 g g(-1)) indicating a good agreement with the predicted value. C. guilliermondii was able to convert vanillin completely after 24 h of fermentation with 94% yield of vanillyl alcohol. CONCLUSIONS: The bioconversion of xylose into xylitol by C. guilliermondii is strongly dependent on the combination of aldehydes and phenolics in the fermentation medium. Vanillin is a source of phenolic compound able to improve xylitol production by yeast. The conversion of vanillin to alcohol vanilyl reveals the potential of this yeast for medium detoxification. (C) 2009 Society of Chemical Industry