Auditory brainstem response classification for threshold detection using estimated evoked potential data: comparison with ensemble averaged data

Auditory brainstem response (ABR) has become a routine clinical tool in neurological and audiological assessment. ABR measurement process with ensemble averaging is very time-consuming and uncomfortable for subjects due to the more repetition of single trials. This condition also restricts the wide usability of ABR in clinical applications. Therefore, the reduction in repetitions has a great importance in ABR measurements. In this study, 488 ABR responses are used for creating two different data sets. The first set is created conventionally by ensemble averaging of 1,024 single trials for each ABR pattern. The second set is obtained from the first estimated 64 single trials of the same records for each ABRs. Estimation is realized by using a nonlinear adaptive filtering algorithm. In classification stage, a powerful classifier integrated with a feature selection algorithm is performed for each data set. In result, the classification performance for estimated ABR data with 64 repetitions is better than the classification performance of the ensemble averaged data with 1,024 repetitions. The proposed system is resulted in an accuracy of 96% for estimated ABRs. So, the proposed system can effectively be used for threshold detection in auditory assessment providing a high accuracy. While the obtained results contribute to the practical ABR usage in clinics, the great significance of it arises from the reduction in repetitions via estimation of ABRs.     

Optimizing area under the ROC curve using semi-supervised learning

Receiver operating characteristic (ROC) analysis is a standard methodology to evaluate the performance of a binary classification system. The area under the ROC curve (AUC) is a performance metric that summarizes how well a classifier separates two classes. Traditional AUC optimization techniques are supervised learning methods that utilize only labeled data (i.e., the true class is known for all data) to train the classifiers. In this work, inspired by semi-supervised and transductive learning, we propose two new AUC optimization algorithms hereby referred to as semi-supervised learning receiver operating characteristic (SSLROC) algorithms, which utilize unlabeled test samples in classifier training to maximize AUC. Unlabeled samples are incorporated into the AUC optimization process, and their ranking relationships to labeled positive and negative training samples are considered as optimization constraints. The introduced test samples will cause the learned decision boundary in a multi-dimensional feature space to adapt not only to the distribution of labeled training data, but also to the distribution of unlabeled test data. We formulate the semi-supervised AUC optimization problem as a semi-definite programming problem based on the margin maximization theory. The proposed methods SSLROC1 (1-norm) and SSLROC2 (2-norm) were evaluated using 34 (determined by power analysis) randomly selected datasets from the University of California, Irvine machine learning repository. Wilcoxon signed rank tests showed that the proposed methods achieved significant improvement compared with state-of-the-art methods. The proposed methods were also applied to a CT colonography dataset for colonic polyp classification and showed promising results.¹     

Determination of optimal hierarchical fuzzy controller parameters according to loading condition with ANN

This paper represents an artificial neural network (ANN) backpropagation algorithm is used to choose best coefficients of hierarchical fuzzy power system stabilizer (HFPSS). PSS is used for stability enhancement of a single machine infinite bus (SMIB) power system. ANN algorithm is used to predict load condition of the power system. And according to the predicted load condition ANN determinates choosing optimal parameters of the hierarchical fuzzy controller (HFC) to achieve better performance. Simulation results are compared with conventional PSS (CPSS) to show the effectiveness of the proposed controller. Also quantitative criterias of measuring performance is computed for 16 loading conditions.     

Classification of mass and normal breast tissue: a convolution neural network classifier with spatial domain and texture images

The authors investigated the classification of regions of interest (ROI's) on mammograms as either mass or normal tissue using a convolution neural network (CNN). A CNN is a backpropagation neural network with two-dimensional (2-D) weight kernels that operate on images. A generalized, fast and stable implementation of the CNN was developed. The input images to the CNN were obtained from the ROI's using two techniques. The first technique employed averaging and subsampling. The second technique employed texture feature extraction methods applied to small subregions inside the ROI. Features computed over different subregions were arranged as texture images, which were subsequently used as CNN inputs. The effects of CNN architecture and texture feature parameters on classification accuracy were studied. Receiver operating characteristic (ROC) methodology was used to evaluate the classification accuracy. A data set consisting of 168 ROIs containing biopsy-proven masses and 504 ROI's containing normal breast tissue was extracted from 168 mammograms by radiologists experienced in mammography. This data set was used for training and testing the CNN. With the best combination of CNN architecture and texture feature parameters, the area under the test ROC curve reached 0.87, which corresponded to a true-positive fraction of 90% at a false positive fraction of 31%. The authors' results demonstrate the feasibility of using a CNN for classification of masses and normal tissue on mammograms.     

Characterization of poly(N‐(hydroxymethyl)methacrylamide‐ATU) hydrogels synthesized by γ radiation

Poly(N‐(hydroxymethyl)methacrylamide‐1‐allyl‐2‐thiourea), (poly(NHMMA‐ATU)) hydrogels were synthesized by γ radiation, using ⁶⁰Co γ source at different radiation doses, to change the porosity and crosslinking density of the hydrogels. The percent of 1‐allyl‐2‐thiourea (ATU) in the monomer mixture before the irradiation was varied between 2.5% and 10.0%, to increase the content of ATU, which was involved in some different applications in the hydrogels. Fourier transform infrared spectroscopy (FTIR), thermogravimetric analysis (TGA), elemental analysis, and the swelling experiments were used to characterize the poly(NHMMA‐ATU) hydrogels synthesized in this study. Characterization results of hydrogels showed that crosslinking density of the hydrogels was increased by the increasing radiation dose and ATU content in the irradiated mixture. Swellability of these hydrogels was found to be high enough to allow the metal ions and biomolecules getting inside the hydrogels to interact with all active groups on/in the hydrogels in the adsorption applications. Equilibrium swelling ratio of the hydrogels at pH 0.5 is at least half of the equilibrium swelling ratio of the hydrogels at pH 7.0. Oscillatory swelling behavior of poly(NHMMA‐ATU) hydrogels between pH 0.5 and pH 7.0 showed that the hydrogels are quite stable at different pH conditions. © 2005 Wiley Periodicals, Inc. J Appl Polym Sci 99: 1657–1664, 2006     

Controllable hydrogen generation by use smart hydrogel reactor containing Ru nano catalyst and magnetic iron nanoparticles

In this study, p(AMPS) hydrogels are synthesized from 2-acrylamido-2-methyl-1-propansulfonic acid (AMPS) via a photo polymerization technique. The hydrogels are used as template for metal nanoparticles and magnetic ferrite nanoparticles, and also as a catalysis vessel in the generation of hydrogen from the hydrolysis of NaBH₄. Approximately 5 nm Ru (0) and 20-30 nm magnetic ferrite particles are generated in situ inside this p(AMPS) hydrogel network and then used as a catalysis medium in hydrogen production by hydrolysis of sodium boron hydride in a basic medium. With an applied external magnetic field, the hydrogel reactor, containing Ru and ferrite magnetic particles, can be removed from the catalysis medium; providing on-demand generation of hydrogen. The effect of various parameters such as the initial concentration of NaBH₄, the amount of catalyst and temperature on the hydrolysis reaction is evaluated. The activation energy for hydrogen production by Ru (0) nanoparticles is found to be 27.5 kJ mol⁻¹; while the activation enthalpy is 30.4 kJ mol⁻¹. The hydrogen generation rate in presence of 5 wt% NaOH and 50 mg p(AMPS)-Ru catalyst is 8.2 L H₂ min⁻¹ g Ru.     

The on demand generation of hydrogen from Co-Ni bimetallic nano catalyst prepared by dual use of hydrogel: As template and as reactor

We report the preparation of metal nanoparticles in various formulations inside p(2-acrylamido-2-methyl-1-propansulfonic acid; p(AMPS)) hydrogels and their utilization as a catalyst in hydrolysis of NaBH₄. The swollen, flexible p(AMPS) network was used for metal ion loading and reduction in situ for the preparation of Co:Ni nanoparticles as bimetallic clusters in various formulation, and Co and Ni bimetallic catalysts as Co + Co, Co + Ni, Ni + Co and Ni + Ni. In addition to utilization of hydrogels as support materials, the p(AMPS)-metal nanoparticle system was used as catalyst to generate hydrogen in the hydrolysis of NaBH₄ with very high yield. Various parameters for the hydrolysis reaction were determined and the activation parameters were calculated. For the first time, inclusion of ferrite magnetic particles to control hydrogen generation on demand by using an externally applied magnetic field to remove the hydrogel-catalyst system from the hydrolysis medium is reported.     

Flower pollination–feedforward neural network for load flow forecasting in smart distribution grid

Neural Computing and Applications - Nature-inspired population-based metaheuristic flower pollination algorithm is proposed in solving load flow forecasting problem in smart distribution grid...     

Functionalization of halloysite nanotubes with polyethyleneimine and various ionic liquid forms with antimicrobial activity

Halloysite nanotube (HNT), a natural clay, was modified with branched polyethyleneimine (PEI) to form PEI-HNT using epichlorohydrin (ECH) as coupling agent, then protonated with HCl to obtain H-PEI-HNTs providing [NH₃]⁺[Cl]⁻ functionality for potential antimicrobial properties. Upon PEI modification, zeta potential value of HNTs was increased to +37.3 mV from −34.5 mV and to +41.1 mV for H-PEI-HNTs. Only 1.87 wt % H-element in HNT was increased to 3.03 wt % upon PEI modification along with newly generated elements of N and C at 2.99 and 9.93 wt %, respectively. Moreover, ionic liquid (IL) forms of HNTs with [NH₃]⁺[N(CN)₂]⁻, [NH₃]⁺[PF₆]⁻ and [NH₃]⁺[BF₄]⁻ functionality were generated via anion exchange of H-PEI-HNTs with sodium dicyanamide (SDC), ammonium hexafluorophosphate (AHFP), and sodium tetrafluoroborate (STFB). The antimicrobial properties of the modified, protonated, and IL forms of HNTs were determined via macro dilution, diffusion and agar screening tests against Escherichia coli ATCC 8739, Pseudomonas aeruginosa ATCC 10145, Bacillus subtilis ATCC 6633, Staphylococcus aureus ATCC 6538 strains, and Candida albicans ATCC 10231 strains. It was found that H-PEI-HNTs possesses potent antimicrobial effect compared with the other forms of HNTs with 2–4 mg mL⁻¹ MIC and 8–16 mg mL⁻¹ MBC values via the macro dilution method. © 2019 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2020 , 137, 48352.