Showing papers by "Sirindhorn International Institute of Technology published in 2011"

A framework for automatic heart sound analysis without segmentation

Abstract: A new framework for heart sound analysis is proposed. One of the most difficult processes in heart sound analysis is segmentation, due to interference form murmurs. Equal number of cardiac cycles were extracted from heart sounds with different heart rates using information from envelopes of autocorrelation functions without the need to label individual fundamental heart sounds (FHS). The complete method consists of envelope detection, calculation of cardiac cycle lengths using auto-correlation of envelope signals, features extraction using discrete wavelet transform, principal component analysis, and classification using neural network bagging predictors. The proposed method was tested on a set of heart sounds obtained from several on-line databases and recorded with an electronic stethoscope. Geometric mean was used as performance index. Average classification performance using ten-fold cross-validation was 0.92 for noise free case, 0.90 under white noise with 10 dB signal-to-noise ratio (SNR), and 0.90 under impulse noise up to 0.3 s duration. The proposed method showed promising results and high noise robustness to a wide range of heart sounds. However, more tests are needed to address any bias that may have been introduced by different sources of heart sounds in the current training set, and to concretely validate the method. Further work include building a new training set recorded from actual patients, then further evaluate the method based on this new training set.

Automatic retinal vessel tortuosity measurement using curvature of improved chain code

Abstract: Measurement of blood vessel tortuosity is a useful capability for automatic ophthalmological diagnostic tools. Screening of Retinopathy of Prematurity (ROP), a disease of eye that affects premature infants, for example, depends crucially on automatic tortuosity evaluation. Quite a few techniques for tortuosity measurement and classification have been proposed, but they do not always match the clinical concept of tortuosity. In this paper, we propose the alternative method of automatic tortuosity measurement for retinal blood vessels that uses the curvature calculated from improved chain code algorithm taking the number of inflection point into account. The tortuosity calculated from the proposed method is independent of the segmentation of vessel tree. Our algorithm can automatically classify the image as tortuous or non-tortuous. The test results are verified against two expert ophthalmologists. For an optimal set of training parameters the prediction is as high as 100% on 18 images.

Analysis of Energy Consumption in Drying Process of Non-Hygroscopic Porous Packed Bed Using a Combined Multi-Feed Microwave-Convective Air and Continuous Belt System (CMCB)

Abstract: In this study, the analysis of energy consumption during the drying of non-hygroscopic porous packed bed by combined multi-feed microwave-convective air and continuous belt system (CMCB) was investigated experimentally By using a combined multi-feed microwave-convective air and continuous belt system drier, the microwave power was generated by means of 12 compressed air-cooled magnetrons of 800 W each that give a maximum of 96 kW The power setting could be adjusted individually in 800 W steps Hot air with the maximum working temperature of 240°C was generated using 24 units of electric heater where the total power capacity is 108 kW Most importantly, this work focused on the investigation of drying phenomena under industrialized microwave processing In this analysis, the effects of the drying time, hot air temperature, porous structure (F-Bed and C-Bed), and location of magnetrons on overall drying kinetics and energy consumption were evaluated in detail The results showed that the overall drying

Fuzzy logic based PWM control and neural controlled-variable estimation of pneumatic artificial muscle actuators

Abstract: Research highlights? Mathematical modeling of pneumatic artificial muscle (PAM) actuators is developed. ? Fuzzy based PWM control is applied to regulate dynamic behaviors of PAM actuators. ? Minimum-time control is embedded in fuzzy rules. ? Neural network is used to estimate inaccessible variables: Position and force. ? Experiments of industrial PAM control are performed under working conditions. There is increasing research to explore both efficient and cost-effective control of pneumatic artificial muscle (PAM) actuators against inherent nonlinear behaviors of the actuators themselves. This paper presents a fuzzy logic based Pulse-Width-Modulation (PWM) control of PAM actuators together with controlled-variable estimation of a neural network. The PAM actuator consists of two main non-linear elements: a PAM and an on/off control valve unit. Dynamic modeling of the PAM actuator is carried out so as to represent a real PAM actuator in simulation of the dynamic behaviors for gaining knowledge in the controller design. The proportional-type fuzzy control law based on a minimum-time control design is proposed to determine the mass flow rate of compressed air in manipulating the controlled variables of the PAM actuators, such as position and force. In circumstances, when the controlled variables are inaccessible, a neural network model is proposed to estimate those variables instead of using direct measurement. The class of PAM actuators available in industry is used as a practical example to show the effectiveness of the proposed methodology in real working conditions. The concept of the proposed knowledge-based control system, which can emulate the reasoning procedures in this work, can be generalized to systematically implement other PAM actuators in real-time control.

Elementary chaotic snap flows

Abstract: Hyperjerk systems with 4th-order derivative of the form x….=f(x…,x¨,x˙,x) have been referred to as snap systems. Five new elementary chaotic snap flows and a generalization of an existing flow are presented through an extensive numerical search. Four of these flows demonstrate elegant simplicity of a single control parameter based on a single nonlinearity of a quadratic, a piecewise-linear or an exponential type. Two others demonstrate elegant simplicity of all unity-in-magnitude parameters based on either a single cubic nonlinearity or three cubic nonlinearities. The chaotic snap flow with a single cubic nonlinearity requires only two terms and can be transformed to its equivalent dynamical form of only five terms which have a single nonlinearity. An advantage is that such a chaotic flow offers only five terms even though the (four) dimension is high. Three of the chaotic snap flows are characterized as conservative systems whilst three others are dissipative systems. Basic dynamical properties are described.

Tensile behavior of restrained expansive mortar and concrete

Abstract: Effect of restraint on the tensile behavior of expansive mortar was investigated by testing specimens with concentrically un-bonded restraint, chemically prestressed mortar (CPM), and chemically prestressed concrete (CPC). The deformation under flexural load is measured by a series of strain gages. In the test of specimens with un-bonded restraint, the restraining steel was removed and the expansion, chemical prestress, and rebound strain are measured before the flexural test. Tensile strength, cracking strain capacity as well as the plastic deformation was quantitatively investigated. The results show that the restrained expansive mortar has unique tensile properties, i.e., larger cracking strain capacity, non-linearity, and more plastic deformation. The effects of bonding between rebar and mortar, curvature, and the presence of coarse aggregate on the tensile properties of CPM and CPC were also experimentally investigated.

Treating swine wastewater by integrating earthworms into constructed wetlands.

Abstract: The aim of this study was to investigate the application of integrating earthworms (Pheretima peguana) into two-stage pilot-scale subsurface-flow constructed wetlands (SFCWs) receiving swine wastewater in terms of their treatment performance, namely organic content, total kjeldahl nitrogen (TKN), and solid reduction as well as the quantity of sludge production. There was a minor difference in terms of removal efficiency according to each parameter when comparing the unit with earthworms to the one without earthworms. Both achieved the TKN, biochemical oxygen demand (BOD), chemical oxygen demand (COD), total volatile suspended solids (TVSS), suspended solids (SS), and total solids (TS) removal by more than 90 %. The earthworms helped in reducing the sludge production on the surface of constructed wetlands 40 % by volume, which resulted in lowering operational costs required to empty and treat the sludge. The plant biomass production was higher in the wetlands without earthworms. Further research could be undertaken in order to effectively apply earthworms inside the wetlands.

Phosphorus recovery from human urine and anaerobically treated wastewater through pH adjustment and chemical precipitation.

Abstract: Increased population growth and food prices have resulted in more demand for fertilizers, especially phosphorus (P), to be used in agriculture and production of food crops. This research investigated the feasibility of P recovery from selected wastewaters in the form of precipitates only with pH adjustment. Human urine and effluent of an anaerobic digester treating a piggery's wastewater were employed to determine appropriate pH conditions for P recovery including solubility of the precipitates in an agricultural soil. From the laboratory experiments, the highest P recovery was found to be 106 mg per one litre of urine at pH 11. Due to its lower P content, the highest P recovery from one litre of the anaerobic digester effluent was 39 mg at pH 9. The X‐ray diffraction analysis of the precipitates identified them to consist of struvite, syn‐NH4MgPO4·6H2O, and other precipitate compounds consisting of CaCO3, NaCl and Mg3Al2(SiO4)3 and P contents of the precipitate samples were found to be 3–7%. When mixed w...

Meaning-preserving skolemization

Abstract: Skolemization is a well-known method for removing existential quantifiers from a logical formula. Although it always yields a satisfiability-preserving transformation step, classical Skolemization in general does not preserve the logical meaning of a source formula. We develop in this paper a theory for extending a space of logical formulas by incorporation of function variables and show how meaning-preserving Skolemization can be achieved in an obtained extended space. A procedure for converting a logical formula into an equivalent one in an extended conjunctive normal form on the extended space is described. This work lays a theoretical foundation for solving logical problems involving existential quantifications based on meaning-preserving formula transformation.

Medical image blob detection with feature stability and KNN classification

Abstract: Object detection methods with feature stability along with the use of KNN algorithm are proposed within this paper. A scale-space tree is constructed based on the blobs that were created from a series of images after blurring. Features and spatial information provides the role in scalespace tree construction. After the process of blob extraction, users determine the type of blobs that were detected within the image by distinguishing classes to create ground truth images. Within the same process, KNN algorithm is applied to distinguish classes of the image's blob to demonstrate its performance.

Automatic extraction of retinal vessels based on gradient orientation analysis

Abstract: Retinal vessel extraction is important for the diagnosis of numerous eye diseases. It plays an important role in automatic retinal disease screening systems. This paper presents an efficient method for the automated analysis of retinal images. Fine anatomical features, such as blood vessels, are detected by analyzing the gradient orientation of the retinal images. The method is independent of image intensity and gradient magnitude; therefore, it performs accurately despite the common problems inherent to the retinal images, such as low contrast and non-uniform illumination. Blood vessels with varying diameters are detected by applying this method at multiple scales. The blood vessel network is then extracted from the detected features by manual thresholding followed by a few simple morphological operations. Based on the binary vessel map obtained, we attempt to evaluate the performance of the proposed algorithm on two publicly available databases (DRIVE and STARE database) of manually labeled images. The receiver operating characteristics (ROC), area under ROC and segmentation accuracy is taken as the performance criteria. The results demonstrate that the proposed method outperforms other unsupervised methods in respect of maximum average accuracy (MAA). The proposed method results in the area under ROC and the accuracy of 0.9037, 0.9358 for DRIVE database 0.9117, 0.9423 for STARE database respectively.

Minimization of the kinematics error for five-axis machining

Abstract: Kinematics of a particular five-axis milling machine can drastically change the machining accuracy. Therefore, the reduction of the kinematics error is an important problem associated with the tool path planning. Our new optimization method employs a closed form of the kinematics error represented as a function of the positions of the cutter contact points. The closed form is derived from the inverse kinematics associated with a particular five-axis machine and obtained through automatic symbolic calculations. The second component of the algorithm is the optimal setup of the part surface on the mounting table employed in an iterative loop with the generation of the cutter contact points. For a prescribed tolerance the proposed optimization allows for substantial reduction in the number of required cutter contact points. The reduction can be significant and may amount to long hours of machining if the machining time at the programmed feed is less than the sampling time of the controller. In turn, when the number of cutter location points is fixed, the error can be substantially reduced. However, this refers to commanded error wherein the dynamics of machine tool are not taken into account. We present an analysis, systematic numerical experiments and results of real cutting (ball nose and flat-end cutters) as an evidence of the efficiency and the accuracy increase produced by the proposed method. We also evaluate the relative contributions of the setup and the point optimization. The method is shown to work with advanced tool path generation techniques proposed earlier such as the adaptive space filling curves. The numerical and machining experiments demonstrate that the proposed procedure outperforms tool paths based on the equi-arclength principle and paths generated by MasterCam 9.

Numerical experiments with cooperating multiple quadratic snakes for road extraction

Abstract: Higher-order active contours or snakes show much promise for the extraction of complex objects from noisy imagery. These models provide an elegant mathematical framework for specifying the desired properties of target objects through energy functionals that can be minimized with standard optimization techniques. However, techniques to allow quadratic snakes to change topology during segmentation have not been fully exploited. Additionally, external forces for improving convergence of quadratic snakes have similarly yet to be explored. In this article, we propose a model that allows multiple quadratic snakes to split, merge, and disappear. Although the separate components of our approach have been introduced elsewhere by Cohen (1991), Xu and Prince (1997), and Rochery et al. (2006), this article is the first comprehensive empirical study of their performance on real-world complex network extraction tasks. We analyze the applicability of the model to road extraction from satellite images that vary in complexity from simple networks to large networks with multiple loops. We also analyze the effects of external forces enhanced by oriented filtering, gradient vector flow fields, and Canny edge detection. In a series of experiments, we found that the multiple cooperating quadratic snake model performs well on complex, noisy images. Our experiments also establish a performance improvement when the proposed quadratic model is coupled with the Canny-based gradient vector flow technique.

Algebraic decoder of multidimensional convolutional code: constructive algorithms for determining syndrome decoder and decoder matrix based on Gröbner basis

Abstract: A representation of an multidimensional (m-D) convolutional encoder is analogous to the representation of a transfer function for a MIMO m-D FIR system. The encoder matrix is usually not square and thus finding its inverse (decoder matrix) typically employs the Moore-Penrose generalized inverse. However, the result may not be FIR (polynomial matrix) even if the generator matrix is a polynomial matrix. In this paper a constructive algorithm for computing the FIR pseudo inverse, based on the usage of Grobner basis is presented along with detailed examples. The result obtained can be parameterized to cover the class of all possible FIR inverses. In addition, by using the computation method of syzygy with the Grobner basis module, the syndrome matrix for a given m-D convolutional encoder is shown. Furthermore, the theory of Grobner basis is applied to solve the algebraic syndrome decoder problems using the maximum likelihood (nearest neighborhood) criteria and the procedure for 2-D convolutional code error correction is proposed. Despite the complication of the decoding process, the proposed method is the only error correcting decoder for multidimensional convolutional code available to date.

CO2 mitigation model of future power plants with integrated carbon capture and storage in Thailand

Abstract: Thailand currently relies on natural gas, coal, and oil for electricity production. Natural gas and oil fields are the plentiful domestic energy resources in the Gulf of Thailand, but are projected to be available only in the next decade. Using domestic natural gas would reduce national energy security. Due to its low price, coal will dominate in electricity generation in Thailand. However, more utilization of coal results in increasing CO2 emissions. To comply with the low-carbon society policy of the Royal Thai Government, carbon dioxide capture and storage (CCS) for the future committed power plants in Thailand is introduced and analysed using the market allocation model, a least-cost energy planning tool. In this study, CO2 mitigation scenarios include integrated CCS to new gas- and coal-based power plants, nuclear power and renewable-based power plants and these were introduced as mitigation options. In energy-demand forecasting, share of coal-based plants is projected to increase from 0.72% in 2006 ...

A haptic and auditory assistive user interface: Helping the blinds on their computer operations

Abstract: In this paper, a study of assistive devices with multi-modal feedback is conducted to evaluate the efficiency of haptic and auditory information towards the users' mouse operations. Haptic feedback, generated by a combination of wheels driven by motors, is provided through the use of the haptic mouse. Meanwhile, audio feedback either in the form of synthesized directional speech or audio signal. Based on these interfaces, a set of experiments are conducted to compare their efficiencies. The measurement criteria used in this experiment are the distance regarding to the target circle in pixels, the operational time for the task in milliseconds, and opinion in term of understandability and comfortability towards each modal of the tested user interfaces in discrete indices. The experimental results show that with the proper modalities of feedback interfaces for the user, the efficiency can be improved by either the reduction in operational time or the increase of accuracy in pointing the target. Furthermore, the justification is also based on the user's satisfaction towards using the device to conduct the predefined cursor movement task, which occasionally is difficult to understand and interpret by the user. For example of the application adopting the proposed interface system, a web browser application is implemented and explained in this paper.

Analytic-based decision analysis tool for employee-job assignments based on competency and job preference

Abstract: This paper discusses an analytic-based decision analysis tool called Employee-Job Assignment System (EJAS) for solving the employee-job assignment problem (EJAP). The EJAS is intended to determine an employee-job assignment (EJA) solution that optimizes the employees’ work efficiency and job satisfaction. Two person-job (P-J) fit criteria are considered: (1) competency-based P-J fit, and (2) preference-based P-J fit. Depending on the decision-maker’s selection, the EJAS determines an EJA solution that maximizes the chosen P-J fit. Additionally, it is capable of generating several near-optimal solutions to assist the decision-maker in selecting a suitable EJA solution. Significance: Effective management of the human resource is an important issue for every business organization. Assigning the right employees to the right jobs can help the organization to increase its productivity and/or enhance its service efficiency. The proposed decision analysis tool enables responsible decision-makers to develop the employee-job assignment solution that considers both competency and job preference; thus, achieving increased productivity and better job satisfaction. Keywords: Employee-job assignment, workforce management, person-job fit, competency, job satisfaction, decision support system

All-pass based IIR multiple notch filter design using Gröbner Basis

Abstract: In this paper, the digital IIR multiple notch filter based on an all-pass filter design of order 2N is considered. The most important step is about the calculation of the exact notch frequencies. The previous method requires solving of non-linear polynomial equation system which can be difficult to tackle especially for the case with higher number of notch frequencies. The analytical result was derived only for the case of N = 2 not for N = 3 and higher, then the main idea proposed in this paper is focused on solving of the non-linear systems for the cases of higher number of notch frequencies, e.g. N ≥ 3, by employing “Grobner Basis” theory which has ability to solve multi-variance polynomial equation systems by using the benefit of “lexicographical orderings” in the polynomial rings to make those systems to be “triangular systems” which can be solved by backward substitution. Although the proposed filter design is a one dimensional but the proposed technique involved multi-variance polynomials. Another advantage is that the Grobner Basis provides symbolic solutions which allow further optimization to satisfy additional constraints such as having minimum group delay.