A nonlinear partial least squares algorithm using quadratic fuzzy inference system

We introduce a new nonlinear partial least squares algorithm ‘Quadratic Fuzzy PLS (QFPLS)’ that combines the outer linear Partial Least Squares (PLS) framework and the Takagi–Sugeno–Kang (TSK) fuzzy inference system. The inner relation between the input and the output PLS score vectors is modeled by a quadratic TSK fuzzy inference system. The performance of the proposed QFPLS method is tested and compared against four other well‐known partial least squares methods (Linear PLS (LPLS), Quadratic PLS (QPLS), Linear Fuzzy PLS (LFPLS), and Neural Network PLS (NNPLS)) on various different types of randomly generated test data. QFPLS outperformed competitors based on two comparison measures: the output variables cumulative per cent variance captured by the PLS latent variables and the root mean‐square error of prediction (RMSEP). Copyright © 2009 John Wiley & Sons, Ltd.     

运用模糊神经网络表达和预测链烷烃pVT性质

采用一种基于遗传算法的新型模糊神经网络方法研究链烷烃类化合物的pVT性质。该方法综合神经网络、遗传算法与模糊系统三种柔性智能计算技术的优点,具有良好的学习能力,不易陷入局部最小区域,学习速度较快,网络知识以模糊语言变量的形式加以表达,易于理解。用分子连接性指数对24种链烷烃化合物结构和pVT数据进行学习,进而预测另外14种未知化合物的pVT性质,较好地揭示出化合物分子结构与pVT性质之间的关系,并给出了良好的关联与预测结果。     

A modified fuzzy C‐means algorithm using scale control spatial information for MRI image segmentation in the presence of noise

The fuzzy C‐means (FCM) algorithm does not fully utilize the spatial information for image segmentation and is sensitive to the presence of noise and intensity inhomogeneity in magnetic resonance imaging (MRI) images. The underlying reason is that using a single fuzzy membership function the FCM algorithm cannot properly represent pattern associations to all clusters. In this paper, we present a modified FCM (mFCM) algorithm by incorporating scale control spatial information for segmentation of MRI images in the presence of high levels of noise and intensity inhomogeneity. The algorithm utilizes scale controlled spatial information from the neighbourhood of each pixel under consideration in the form of a probability function. Using this probability function, a local membership function is introduced for each pixel. Finally, new clustering centre and weighted joint membership functions are introduced based on the local membership and global membership functions. The resulting mFCM algorithm is robust to the noise and intensity inhomogeneity in MRI image data and thereby improves the segmentation results. The experimental results on a synthetic image, four volumes of simulated and one volume of real‐patient MRI brain images show that the mFCM algorithm outperforms k‐means, FCM and some other recently proposed FCM‐based algorithms for image segmentation in terms of qualitative and quantitative studies such as cluster validity functions, segmentation accuracy and tissue segmentation accuracy. Copyright © 2015 John Wiley & Sons, Ltd.     

Comparative analysis of adaptive neuro-fuzzy inference system (ANFIS) and RSRM models to predict DBP (trihalomethanes) levels in the water treatment plant

Since disinfection by-products are a growing concern, it is important to know their quantities in water treatment plants before they are released to the public. As a result, there is a requirement for constant monitoring of disinfection by-products (DBPs), which can have major consequences for human health and productivity. Consequently, in previous studies, several models for predicting disinfection byproduct formation in drinking water have been developed which were either linear/log-linear, hybrid or neural network (radial basis function). In this study, an adaptive neuro-fuzzy inference system (ANFIS) is proposed for predicting trihalomethane levels in real distribution systems. To train and verify the proposed model, 24 sets of data were used, including THMs levels (TCM, BDCM, DBCM and T-THM levels) and five parameters (pH, temperature, UVA₂₅₄, residual chlorine, and dissolved organic carbon). As compared to response surface modeling (RSRM) coefficient of determination, R² is between 0.727 < R² < 0.886, average absolute deviation, AAD = 4.07–10.99 %), MAE = 0.01 – 0.978, and RMSE = 0.017 – 1.449. Further, ANFIS for THMs (T-THMs, TCM, BDCM, and DBCM) prediction consistently show higher regression coefficients between 0.956 < R² < 0.989, average absolute deviation, AAD = 0.350 – 1.977 %), MAE = 0.002 – 0.133, and RMSE = 0.007 – 0.401, Consequently, based on the statistical indices obtained, ANFIS, on the other hand, proved to be effective for predicting the formation of THMs, and thus allowed improved DBPs monitoring in water treatment systems.     

Electrochemical analysis of -penicillamine using a boron-doped diamond thin film electrode applied to flow injection system

The electroanalysis of -penicillamine in 0.1 phosphate buffer (pH 7) was studied at a boron-doped diamond thin film (BDD) electrode using cyclic voltammetry as a function of concentration of analyte and pH of analyte solution. Comparison experiments were performing using a glassy carbon (GC) electrode. The BDD electrode exhibited a well-resolved and irreversible oxidation voltammogram, but the GC electrode provided only an ill-defined response. The BDD electrode provided a linear dynamic range from 0.5 to 10 mM and a detection limit of 25 μM (S/B≥3) in voltammetric measurement. It was also found that the peak potentials were decreased when the pH of the analyte solution was increased. In addition, penicillamine has been studied by hydrodynamic voltammetry and flow injection analysis with amperometric detection using the BDD electrode. The flow injection analysis results at the diamond electrode indicated a linear dynamic range from 0.5 to 50 μM and a detection limit of 10 nM (S/N≈4). The proposed method was applied to determine -penicillamine in dosage form (capsules), the results obtained in the recovery study (255±2.50 mg per tablet) were comparable to those labeled (250 mg per tablet).     

Development and analysis of cellulose nanocomposite films with in situ generated silver nanoparticles using tamarind nut powder as a reducing agent

Cellulose/Tamarind nut powder (TNP)/Silver nanoparticles (AgNPs) nanocomposites were prepared by in situ generation of AgNPs using regeneration method, followed by solution casting method. In this, TNP was used as a reducing agent. These nanocomposites were characterized using FT-IR spectroscopy, XRD and SEM and studied their mechanical properties and antibacterial activity for medical and packing applications. The FT-IR spectral studies revealed the involvement of functional groups – Polyphenols, Flavonoids and –OH in the process of reducing the metal salts into metal nanoparticles. These nanocomposites showed good antibacterial activity against five bacteria. Improved mechanical properties with good antibacterial activities make these composites suitable for medical, food and packaging applications.     

Quantitative analysis of near surfaces three‐dimensional orientation of polymer chains in PET and PEN films using polarized ATR FTIR spectroscopy

Polarized attenuated total reflection (ATR) Fourier transform infrared (FTIR) spectroscopy was utilized to characterize the three‐dimensional orientation of polymer chains near highly anisotropic surfaces generated by uniaxial drawing. A versatile method was proposed to analyze the molecular orientation of the polymers by combining the experimental refractive indices and optimized contact pressure by an anvil for solving the optical contact problem. This method is effective even when changes in the molecular orientation along the thickness direction caused by drawing are remarkable. In addition, this method enables quantitative comparison of the molecular orientation among different polymers in the same coordinate system. From the molecular orientation analysis of poly (ethylene terephthalate) (PET) and poly (ethylene naphthalate) (PEN), it was revealed that this method has a broader range of applications with high accuracy in estimating the molecular orientation of polymers compared with the conventional methods. The significant changes in the molecular orientation caused by uniaxial and biaxial drawing of PET and PEN films were quantitatively analyzed, and the reasons for the significant in‐plane orientation of PEN chains on the film plane are discussed. In addition, the difference in the molecular orientation between both sides of the films was also demonstrated. © 2010 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 48: 870–879, 2010     

A flow injection method for the analysis of tetracycline antibiotics in pharmaceutical formulations using electrochemical detection at anodized boron-doped diamond thin film electrode

A method using flow injection (FI) with amperometric detection at anodized boron-doped diamond (BDD) thin films has been developed and applied for the determination of tetracycline antibiotics (tetracycline, chlortetracycline, oxytetracycline and doxycycline). The electrochemical oxidation of the tetracycline antibiotics was studied at various carbon electrodes including glassy carbon (GC), as-deposited BDD and anodized BDD electrodes using cyclic voltammetry. The anodized BDD electrode exhibited well-defined irreversible cyclic voltammograms for the oxidation of tetracycline antibiotics with the highest current signals compared to the as-deposited BDD and glassy carbon electrodes. Low detection limit of 10 nM (signal-to-noise RATIO = 3) was achieved for each drug when using flow injection analysis with amperometric detection at anodized BDD electrodes. Linear calibrations were obtained from 0.1 to 50 mM for tetracycline and 0.5–50 mM for chlortetracycline, oxytetracycline and doxycycline. The proposed method has been successfully applied to determine the tetracycline antibiotics in some drug formulations. The results obtained in percent found (99.50–103.01%) were comparable to dose labeled.     

Determination of Rh, Pd and Pt in urine samples using a pre-concentration sequential injection analysis system coupled to a quadrupole-inductively coupled plasma-mass spectrometer

The proposed flow system was developed in order to minimize the drawbacks related to the PGEs determination by quadrupole-inductively coupled plasma-mass spectrometry (Q-ICP-MS). It was intended not only to lower the limits of detection (LODs) but also to eliminate the interferences originating from some atomic and molecular ions produced in the argon plasma. This was accomplished by means of an on-line sample clean-up/pre-concentration step, using a chelating resin (Metalfix™ Chelamine™) in which Rh, Pd and Pt were preferably retained when compared with the interfering species.

The results obtained by using the developed flow system in the analysis of urine samples are presented. With a sampling rate of 9 samples h⁻¹ (i.e., 27 determinations) and a sample consumption of ca. 10 mL, the developed flow system allowed linear calibration plots up to 100 ng L⁻¹ with detection limits of 1.2 ng L⁻¹ (Rh), 0.4 ng L⁻¹ (Pd) and 0.9 ng L⁻¹ (Pt). Repeatability studies showed good precision (R.S.D.%, n = 5): 3.7% (Rh); 2.6% (Pd) and 2.4% (Pt), for 10 ng L⁻¹; 2.4% (Rh); 1.4% (Pd) and 1.9% (Pt), for 50 ng L⁻¹; and 1.3% (Rh); 0.58% (Pd) and 0.62% (Pt), for 100 ng L⁻¹. By spiking human urine samples, recovery tests were performed, and the values obtained ranged between 89% and 105% (Rh); 90% and 104% (Pd); and 93% and 105% (Pt).     

Highly sensitive and quantitative analysis of polyeptides using a new gradient system based on an abrupt change in adsorption of polypeptide to the reversed-phase column packing

During a study of 100 microL aliquots of urocortin containing various acetonitrile contents, we hypothesized that a change in the acetonitrile content in the solution across a specific content of acetonitrile (critical threshold) causes an abrupt change in adsorption capacity to the column packing. Circular dichroism measurements suggest that the conformational change induced by acetonitrile in the solution causes the abrupt change in adsorption capacity, and this solvent-induced conformational change is reversible across the critical threshold. This hypothesis can apply to various polypeptides with molecular weights range from 1007 to 6789 and to other organic solvents. A new gradient system utilizing the instant recovery of the adsorption capacity across the critical threshold was designed, and applied to the analysis of a 100 microL aliquot of various polypeptide solutions. The results suggest that use of a solution containing organic solvents more than the critical threshold allows successful dilution of polypeptides up to picomolar concentration range without any loss due to its adsorption during handling procedures and loading onto the LC system, and that a new gradient system enables quantitative analysis of polypeptides at picomolar concentrations in such solutions.