Mohd Nasir Taib

Extending the response range of an optical fibre pH sensor using an artificial neural network

Abstract An artificial neural network (ANN) has been applied for the analysis of the response of an optical fibre pH sensor. The optrode was based on a 3,4,5,6-tetrabromophenol sulphonephthalein indicator, which was covalently bound onto aminopropyl glass beads, then packed at the tip of a bifurcated fibre-optic bundle. A three layer feed forward network was used and network training was performed using the recursive prediction error (RPE) algorithm. It was found that an optimised network with 13 hidden neurons was highly accurate in predicting the response of the optical pH sensor, with the worst interpolation error of 0.08 pH for test data set and 0.07 pH for measuring unknown buffer solutions. Overall, the application of ANN enabled the extension of the useful pH response range of the sensor from its narrow linear range (pH 5.0–7.25) to the full calibration response (pH 2.51–9.76).

Extending the range of a fibre-optic relative-humidity sensor

Abstract A new signal-processing method to extend the linear operational range of an optical-fibre humidity sensor is presented in this study. The sensor is based on a Nafion-crystal violet complex immobilized on a glass substrate. Low-cost plastic optical fibres are employed as light guides to direct light from a tungsten halogen source to the sensor and from the sensor to a CCD-based spectrometer. Generated spectra for varying relative-humidity levels are analysed using artificial neural networks. Sensor measurements at wavelengths corresponding to the red, orange, yellow and NIR LEDs are used for the artificial neural network input. This study has shown that the artificial neural networks successfully extend the linear response range of the fibre-optic relative-humidity sensor from the 40–55% humidity range previously recorded to a nominal range of 40–82%. The use of LED-compatible wavelengths shows that the sensor can be readily adapted for use with low-cost solid-state instrumentation.

Optimisation of the range of an optical fibre pH sensor using feed-forward artificial neural network

Abstract A broad range of optical fibre pH sensor based on immobilised bromophenol blue (BPB) immobilised on hydrophobic organic polymers Amberlite XAD 7 is presented in this paper. The reflectance spectra of the immobilised bromophenol blue were measured by using an optical fibre spectrophotometer. A back-propagation (BP) artificial neural network (ANN) model was used to analyse the optode response. The results showed that the use of ANN technique was very effective in broadening the limited dynamic response of the pH sensor (pH 2.00–5.00) to an extensive calibration response (pH 2.00–12.00). A network with 11 neurons in the hidden layer was tremendously accurate in predicting the response of the optical fibre pH sensor with an average error 0.02 pH for measuring unidentified buffer solution.

A novel dengue fever (DF) and dengue haemorrhagic fever (DHF) analysis using artificial neural network (ANN)

Dengue fever (DF) is an acute febrile viral disease frequently presented with headache, bone or joint and muscular pains, and rash. A significant percentage of DF patients develop a more severe form of disease, known as dengue haemorrhagic fever (DHF). DHF is the complication of DF. The main pathophysiology of DHF is the development of plasma leakage from the capillary, resulting in haemoconcentration, ascites, and pleural effusion that may lead to shock following defervescence of fever. Therefore, accurate prediction of the day of defervescence of fever is critical for clinician to decide on patient management strategy. To date, no known literature describes of any attempt to predict the day of defervescence of fever in DF patients. This paper describes a non-invasive prediction system for predicting the day of defervescence of fever in dengue patients using artificial neural network. The developed system bases its prediction solely on the clinical symptoms and signs and uses the multilayer feed-forward neural networks (MFNN). The results show that the proposed system is able to predict the day of defervescence in dengue patients with 90% prediction accuracy.

Multichannel calibration technique for optical-fibre chemical sensor using artificial neural network

Abstract A multilayer feed-forward artificial neural network has been utilized to model the input-output data of an optical-fibre pH sensor. An experiment to generate the optical response spectrum and network application is described. Responses from several wavelengths of the spectrum are used as the input data for the neural network, which enables the measurement of pH in the range 1.60–10.18, with maximum error of about 0.3. The neural network tested with noisy data produces an average error of 0.2 pH units for additive noise levels of up to 13% of the input signal.

RUBBER TIRE DUST-RICE HUSK PYRAMIDAL MICROWAVE ABSORBER

Rubber tire dust-rice husk is an innovation in improving the design of pyramidal microwave absorbers to be used in radio frequency (RF) anechoic chambers. An RF anechoic chamber is a shielded room covered with absorbers to eliminate unwanted re∞ection signals. To design the pyramidal microwave absorber, rice husk will be added to rubber tire dust since the study shows that both have high percentages of carbon. This innovative material combination will be investigated to determine the best re∞ectivity or re∞ection loss performance of pyramidal microwave absorbers. Carbon is the most important element that must be in the absorber in order to help the absorption of unwanted microwave signals. In the commercial

Setup and results of pyramidal microwave absorbers using rice husks

Agricultural wastes are considered not useful and are commonly dumped or burned after crop harvesting. Rice husks from paddy (Oryza sativa) are example of agricultural wastes. Rice husks have been investigated as the material for the pyramidal microwave absorbers. The setup for the fabrication and measurement of the rice husks pyramidal microwave absorbers are discussed. An 8£8 array of pyramidal microwave absorber using the rice husks-polyester-MEKP mixture has been designed and fabricated. There are four main stages in this work: the collection of the raw rice husks materials, the mould fabrication, the pyramidal microwave absorber fabrication and the experiments performed to determine the re∞ection loss performance of the rice husks pyramidal microwave absorbers. Experimental

A novel approach to classify risk in dengue hemorrhagic fever (DHF) using bioelectrical impedance analysis (BIA)

This paper introduces a novel approach to classify the risk in dengue hemorrhagic fever (DHF) patients using the bioelectrical impedance analysis (BIA) technique. This in vivo technique involves the application of a small average constant current of less than 1 mA at a single frequency of 50 kHz through the human body, and measurement of the bodys bioelectrical resistance (R), phase angle (/spl alpha/), body capacitance (BC) and capacitive reactance (X/sub c/) via four surface electrodes. BIA measurements have been conducted on 184 (97 males and 87 females) serological confirmed dengue patients during their hospitalization in University Kebangsaan Malaysia Hospital, Malaysia. The patients included in the study were DHF I-IV according to World Health Organization criteria. Univariate analysis of variance is used for assessing the relationship between gender and group with the bioelectrical tissue conductivity (BETC) parameters. Experimental findings show that BETC, as reflected by reactance, is the key determinant indicator for classifying risk category in the DHF patients. Hence, this novel approach of the BIA technique can provide a rapid, noninvasive, and promising method for classifying and evaluating the risk of the DHF patients.

EEG-based Stress Features Using Spectral Centroids Technique and k-Nearest Neighbor Classifier

This paper presents the combination of electroencephalogram (EEG) power spectrum ratio and Spectral Centroids techniques to extract unique features for human stress from EEG signals. The combination of these techniques was able to improve the k-NN (k-Nearest Neighbor) clasifier accuracy to detect and classify human stress from two cognitive states, Close-eye (CE) and Open-eye (OE). The EEG power spectrum in term of Energy Spectral Density (ESD) for each frequency bands (Delta, Theta, Alpha and Beta) was calculated. The ratio of EEG power spectrum and the average value of Spectral Centroids were selected as features to k-Nearest Neighbor (k-NN). The training and testing of the classifier were evaluated at 50:50 ratios and 70:30 ratios. The results showed that the combination of EEG power spectrum and Spectral Centroids techniques with the training and testing of k-NN set at 70:30 able to detect and classify the unique features for human stress at 88.89% accuracy.

Evaluation of human stress using EEG Power Spectrum

This paper presents an evaluation conducted between human stress questionnaires with EEG Power Spectrum of Beta and Alpha band. Cohens Perceived Stress Scale (PSS) was used as stress questionnaires to evaluate human stress. The EEG recording of 13 volunteers were carried out immediately after them answering the stress questionnaires. The scores from the stress questionnaires were calculated and used to figure out its relationship with the ratio of EEG Beta and Alpha band power. The results of the study showed the PSS was negatively correlated with the ratio of EEG Power Spectrum. Besides, the study suggested that it was feasible to use PSS and the ratio of EEG Power Spectrum to determine human stress.