Sazali Yaacob
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Sazali Yaacob has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Mohd Hafiz Fazalul Rahiman
Faculty of Electrical Engineering & Technology Universiti Malaysia Perlis, Pauh Putra Campus 02600 Arau, Perlis, Malaysia
3 shared publications - Muthusamy Hariharan
Department of Electronics Engineering, National Institute of Technology Uttarakhand, Srinagar Garhwal, India
3 shared publications - Ruzairi Abdul Rahim
Universiti Teknologi Malaysia
3 shared publications - Zulkarnay Zakaria
Universiti Malaysia Perlis
3 shared publications - Muhammad Saiful Badri Mansor
Process Tomography & Instrumentation Research Group (Protom-i), Infocomm Research Alliance, Control and Mechatronic Engineering Department, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Johor, Malaysia
2 shared publications Nor Muzakkir Nor Ayob
Process Tomography and Instrumentation Engineering Research Group (PROTOM-i), Infocomm Research Alliance, Faculty of Electrical Engineering, Universiti Teknologi Malaysia,UTM Skudai, Johor 81310, Malaysia
2 shared publications
Latest Publications
- 1 April 2017
- Published by American Scientific Publishers in Journal of Medical Imaging and Health Informatics
- Vol. 7 (2) , 305-315
Hybrid PSOGSA Algorithm and Nonlinear Features for Prediction Body Mass Index (BMI) from Speech Signals
Obesity is considered as a major cause of some serious diseases such as diabetes and cardiovascular diseases and it has become epidemic. In the field of medicine and health, the usage of body mass index (BMI) is very popular. There are some limitations in using BMI in determining the repartition of body fat due to the fact that it is a skewed measurement and does not take fat into account. The researchers have recently started focusing on exploring the association between speech recognition and BMI. In this paper, a new method to predict BMI status via speech signal (normal, obese and overweight) using wavelet packet transform (WPT) and nonlinear entropy features is proposed. In this work, the recorded speech signals (/ah/sounds) are decomposed up to levels five using WPT. Energy, two short-term cepstral based feature extraction methods namely MFCC, LPCC and nonlinear entropy features like Sample entropy, Tsallis, fuzzy, Approximate, Rényi and Shannon are extracted from the wavelet packet coefficients. Furthermore, Particle swarm optimization (PSO), gravitational search algorithm (GSA) and PSOGSA Hybrid optimization were used/applied to further minimize the dimension of feature space and identify the most relevant feature subset, which has higher discrimination ability to distinguish different BMI status. ELM (Extreme learning machine) was used to classify BMI status. The experimental results reveal that the combination of WPT, nonlinear entropy features; feature selection methods and classification give a maximum classification accuracy of 98.65% for prediction of BMI status (97.91% for normal, 98.86% for obese and 96.96% for overweight).
- 1 April 2017
- Published by American Scientific Publishers in Journal of Medical Imaging and Health Informatics
- Vol. 7 (2) , 305-315
Hybrid PSOGSA Algorithm and Nonlinear Features for Prediction Body Mass Index (BMI) from Speech Signals
Obesity is considered as a major cause of some serious diseases such as diabetes and cardiovascular diseases and it has become epidemic. In the field of medicine and health, the usage of body mass index (BMI) is very popular. There are some limitations in using BMI in determining the repartition of body fat due to the fact that it is a skewed measurement and does not take fat into account. The researchers have recently started focusing on exploring the association between speech recognition and BMI. In this paper, a new method to predict BMI status via speech signal (normal, obese and overweight) using wavelet packet transform (WPT) and nonlinear entropy features is proposed. In this work, the recorded speech signals (/ah/sounds) are decomposed up to levels five using WPT. Energy, two short-term cepstral based feature extraction methods namely MFCC, LPCC and nonlinear entropy features like Sample entropy, Tsallis, fuzzy, Approximate, Rényi and Shannon are extracted from the wavelet packet coefficients. Furthermore, Particle swarm optimization (PSO), gravitational search algorithm (GSA) and PSOGSA Hybrid optimization were used/applied to further minimize the dimension of feature space and identify the most relevant feature subset, which has higher discrimination ability to distinguish different BMI status. ELM (Extreme learning machine) was used to classify BMI status. The experimental results reveal that the combination of WPT, nonlinear entropy features; feature selection methods and classification give a maximum classification accuracy of 98.65% for prediction of BMI status (97.91% for normal, 98.86% for obese and 96.96% for overweight).
- 15 September 2013
- Published by Penerbit UTM Press in Jurnal Teknologi (Sciences & Engineering)
- Vol. 64 (5)
Magnetic Induction Tomography: Receiver Circuit and Its Design Criteria
This paper discusses the receiver circuit and criteria for component selection towards the application of a real-time magnetic induction tomography system. Component selection plays an important role since image reconstruction of the object of interest with high quality and at a higher frame rate cannot be achieved without the right parameter criteria. The demands for a high quality imaging system have recently been increasing, especially in industrial processes involving dynamic movement, thus this paper may provide valuable information on better magnetic induction tomography system implementation for industrial processes and biomedical imaging through the use of a coil as a transmitter and also a receiver. The linear back projection algorithm has been employed in this system and has proven capable of identifying the location and size of the object based on the reconstructed images.
- 15 September 2013
- Published by Penerbit UTM Press in Jurnal Teknologi (Sciences & Engineering)
- Vol. 64 (5)
Magnetic Induction Tomography: Receiver Circuit and Its Design Criteria
This paper discusses the receiver circuit and criteria for component selection towards the application of a real-time magnetic induction tomography system. Component selection plays an important role since image reconstruction of the object of interest with high quality and at a higher frame rate cannot be achieved without the right parameter criteria. The demands for a high quality imaging system have recently been increasing, especially in industrial processes involving dynamic movement, thus this paper may provide valuable information on better magnetic induction tomography system implementation for industrial processes and biomedical imaging through the use of a coil as a transmitter and also a receiver. The linear back projection algorithm has been employed in this system and has proven capable of identifying the location and size of the object based on the reconstructed images.
- Open Access
- 1 January 2013
- Published by Springer Nature in BioMedical Engineering OnLine
- Vol. 12 (1) , 44
Classification of emotional states from electrocardiogram signals: a non-linear approach based on hurst
Identifying the emotional state is helpful in applications involving patients with autism and other intellectual disabilities; computer-based training, human computer interaction etc. Electrocardiogram (ECG) signals, being an activity of the autonomous nervous system (ANS), reflect the underlying true emotional state of a person. However, the performance of various methods developed so far lacks accuracy, and more robust methods need to be developed to identify the emotional pattern associated with ECG signals. Emotional ECG data was obtained from sixty participants by inducing the six basic emotional states (happiness, sadness, fear, disgust, surprise and neutral) using audio-visual stimuli. The non-linear feature ‘Hurst’ was computed using Rescaled Range Statistics (RRS) and Finite Variance Scaling (FVS) methods. New Hurst features were proposed by combining the existing RRS and FVS methods with Higher Order Statistics (HOS). The features were then classified using four classifiers – Bayesian Classifier, Regression Tree, K- nearest neighbor and Fuzzy K-nearest neighbor. Seventy percent of the features were used for training and thirty percent for testing the algorithm. Analysis of Variance (ANOVA) conveyed that Hurst and the proposed features were statistically significant (p < 0.001). Hurst computed using RRS and FVS methods showed similar classification accuracy. The features obtained by combining FVS and HOS performed better with a maximum accuracy of 92.87% and 76.45% for classifying the six emotional states using random and subject independent validation respectively. The results indicate that the combination of non-linear analysis and HOS tend to capture the finer emotional changes that can be seen in healthy ECG data. This work can be further fine tuned to develop a real time system.
- 1 January 2013
- Published by Springer Nature in BioMedical Engineering OnLine
- Vol. 12 (1) , 44
Open AccessClassification of emotional states from electrocardiogram signals: a non-linear approach based on hurst
Identifying the emotional state is helpful in applications involving patients with autism and other intellectual disabilities; computer-based training, human computer interaction etc. Electrocardiogram (ECG) signals, being an activity of the autonomous nervous system (ANS), reflect the underlying true emotional state of a person. However, the performance of various methods developed so far lacks accuracy, and more robust methods need to be developed to identify the emotional pattern associated with ECG signals. Emotional ECG data was obtained from sixty participants by inducing the six basic emotional states (happiness, sadness, fear, disgust, surprise and neutral) using audio-visual stimuli. The non-linear feature ‘Hurst’ was computed using Rescaled Range Statistics (RRS) and Finite Variance Scaling (FVS) methods. New Hurst features were proposed by combining the existing RRS and FVS methods with Higher Order Statistics (HOS). The features were then classified using four classifiers – Bayesian Classifier, Regression Tree, K- nearest neighbor and Fuzzy K-nearest neighbor. Seventy percent of the features were used for training and thirty percent for testing the algorithm. Analysis of Variance (ANOVA) conveyed that Hurst and the proposed features were statistically significant (p < 0.001). Hurst computed using RRS and FVS methods showed similar classification accuracy. The features obtained by combining FVS and HOS performed better with a maximum accuracy of 92.87% and 76.45% for classifying the six emotional states using random and subject independent validation respectively. The results indicate that the combination of non-linear analysis and HOS tend to capture the finer emotional changes that can be seen in healthy ECG data. This work can be further fine tuned to develop a real time system.
- 1 August 2012
- Published by Elsevier in Expert Systems with Applications
- Vol. 39 (10) , 9515-9523
Infant cry classification to identify asphyxia using time-frequency analysis and radial basis neural networks
- 1 August 2012
- Published by Elsevier in Expert Systems with Applications
- Vol. 39 (10) , 9515-9523
Infant cry classification to identify asphyxia using time-frequency analysis and radial basis neural networks
- Open Access
Advancements in Transmitters and Sensors for Biological Tissue Imaging in Magnetic Induction Tomography
Magnetic Induction Tomography (MIT), which is also known as Electromagnetic Tomography (EMT) or Mutual Inductance Tomography, is among the imaging modalities of interest to many researchers around the world. This noninvasive modality applies an electromagnetic field and is sensitive to all three passive electromagnetic properties of a material that are conductivity, permittivity and permeability. MIT is categorized under the passive imaging family with an electrodeless technique through the use of excitation coils to induce an electromagnetic field in the material, which is then measured at the receiving side by sensors. The aim of this review is to discuss the challenges of the MIT technique and summarize the recent advancements in the transmitters and sensors, with a focus on applications in biological tissue imaging. It is hoped that this review will provide some valuable information on the MIT for those who have interest in this modality. The need of this knowledge may speed up the process of adopted of MIT as a medical imaging technology.
Open AccessAdvancements in Transmitters and Sensors for Biological Tissue Imaging in Magnetic Induction Tomography
Magnetic Induction Tomography (MIT), which is also known as Electromagnetic Tomography (EMT) or Mutual Inductance Tomography, is among the imaging modalities of interest to many researchers around the world. This noninvasive modality applies an electromagnetic field and is sensitive to all three passive electromagnetic properties of a material that are conductivity, permittivity and permeability. MIT is categorized under the passive imaging family with an electrodeless technique through the use of excitation coils to induce an electromagnetic field in the material, which is then measured at the receiving side by sensors. The aim of this review is to discuss the challenges of the MIT technique and summarize the recent advancements in the transmitters and sensors, with a focus on applications in biological tissue imaging. It is hoped that this review will provide some valuable information on the MIT for those who have interest in this modality. The need of this knowledge may speed up the process of adopted of MIT as a medical imaging technology.
- 21 March 2012
- Published by Penerbit UTM Press in Jurnal Teknologi (Sciences & Engineering)
- Vol. 55 (1) , 11–14
Initial Development On Magnetic Induction Tomography Imaging
Sistem pengimejan yang tidak intrusif telah banyak menarik perhatian di dalam banyak aplikasi seperti pemprosesan, industri dan perubatan. Teknik elektrikal khususnya, terbukti menjadi instrumen pengimejan berkos rendah dengan kemampuan resolusi rendah tetapi memadai untuk pengimejan pemprosesan. Tomografi Induktansi Magnetik (TIM) adalah kaedah tomografi elektrikal berdasarkan penggunaan bacaan induktansi untuk memantau distribusi bahan konduksi elektrik dan magnetik dalam kawasan pemerhatian. Karya penyelidikan ini adalah berdasarkan pembangunan perisian untuk kegunaan pengimejan tomografi induktansi magnetik. Hasil penyelidikan tertumpu untuk menghasilkan peta sensitiviti yang diperlukan untuk rekonstruksi semula imej tomografi dari taburan elektromagnetik. Keputusan awal mengenai kajian simulatif telah menunjukkan hasil yang positif iaitu peta sensitiviti yang dihasilkan adalah bersesuaian untuk digunakan dalam pengimejan konduktiviti elektrik. Kata kunci: Tomografi induktansi magnetik; pengimejan tidak-intrusif; induktansi; peta sensitiviti Non–intrusive imaging systems have always been of much interest for use in many applications such as process, industrial and medical. Electrical techniques, in particular, are proving to be an inexpensive imaging instrument with low but sufficient resolution capability on imaging the internal distributions processes. Magnetic Induction Tomography (MIT) is an electrical tomographic method based on the use of inductance measurement for monitoring the distribution of electrically conductive and magnetically permeable material within the sensing area. This paper details the fundamental investigation of the software development for magnetic induction tomography imaging. Research works are concentrated on generating the sensitivity maps needed to reconstruct tomographic images of the electromagnetic distribution. Initial results on the simulative studies have shown acceptable result for using the generated sensitivity map for imaging electrical conductivities. Key words: Magnetic induction tomography; non-intrusive imaging; inductance; sensitivity map
- 21 March 2012
- Published by Penerbit UTM Press in Jurnal Teknologi (Sciences & Engineering)
- Vol. 55 (1) , 11–14
Initial Development On Magnetic Induction Tomography Imaging
Sistem pengimejan yang tidak intrusif telah banyak menarik perhatian di dalam banyak aplikasi seperti pemprosesan, industri dan perubatan. Teknik elektrikal khususnya, terbukti menjadi instrumen pengimejan berkos rendah dengan kemampuan resolusi rendah tetapi memadai untuk pengimejan pemprosesan. Tomografi Induktansi Magnetik (TIM) adalah kaedah tomografi elektrikal berdasarkan penggunaan bacaan induktansi untuk memantau distribusi bahan konduksi elektrik dan magnetik dalam kawasan pemerhatian. Karya penyelidikan ini adalah berdasarkan pembangunan perisian untuk kegunaan pengimejan tomografi induktansi magnetik. Hasil penyelidikan tertumpu untuk menghasilkan peta sensitiviti yang diperlukan untuk rekonstruksi semula imej tomografi dari taburan elektromagnetik. Keputusan awal mengenai kajian simulatif telah menunjukkan hasil yang positif iaitu peta sensitiviti yang dihasilkan adalah bersesuaian untuk digunakan dalam pengimejan konduktiviti elektrik. Kata kunci: Tomografi induktansi magnetik; pengimejan tidak-intrusif; induktansi; peta sensitiviti Non–intrusive imaging systems have always been of much interest for use in many applications such as process, industrial and medical. Electrical techniques, in particular, are proving to be an inexpensive imaging instrument with low but sufficient resolution capability on imaging the internal distributions processes. Magnetic Induction Tomography (MIT) is an electrical tomographic method based on the use of inductance measurement for monitoring the distribution of electrically conductive and magnetically permeable material within the sensing area. This paper details the fundamental investigation of the software development for magnetic induction tomography imaging. Research works are concentrated on generating the sensitivity maps needed to reconstruct tomographic images of the electromagnetic distribution. Initial results on the simulative studies have shown acceptable result for using the generated sensitivity map for imaging electrical conductivities. Key words: Magnetic induction tomography; non-intrusive imaging; inductance; sensitivity map
- 25 June 2011
- Published by Elsevier in Expert Systems with Applications
- Vol. 38 (12) , 15377-15382
Pathological infant cry analysis using wavelet packet transform and probabilistic neural network
- 25 June 2011
- Published by Elsevier in Expert Systems with Applications
- Vol. 38 (12) , 15377-15382
Pathological infant cry analysis using wavelet packet transform and probabilistic neural network
- 16 September 2010
- Published by Springer Nature in Journal of Medical Systems
- Vol. 36 (3) , 1309-1315
Analysis of Infant Cry Through Weighted Linear Prediction Cepstral Coefficients and Probabilistic Neural Network
Acoustic analysis of infant cry signals has been proven to be an excellent tool in the area of automatic detection of pathological status of an infant. This paper investigates the application of parameter weighting for linear prediction cepstral coefficients (LPCCs) to provide the robust representation of infant cry signals. Three classes of infant cry signals were considered such as normal cry signals, cry signals from deaf babies and babies with asphyxia. A Probabilistic Neural Network (PNN) is suggested to classify the infant cry signals into normal and pathological cries. PNN is trained with different spread factor or smoothing parameter to obtain better classification accuracy. The experimental results demonstrate that the suggested features and classification algorithms give very promising classification accuracy of above 98% and it expounds that the suggested method can be used to help medical professionals for diagnosing pathological status of an infant from cry signals.
- 16 September 2010
- Published by Springer Nature in Journal of Medical Systems
- Vol. 36 (3) , 1309-1315
Analysis of Infant Cry Through Weighted Linear Prediction Cepstral Coefficients and Probabilistic Neural Network
Acoustic analysis of infant cry signals has been proven to be an excellent tool in the area of automatic detection of pathological status of an infant. This paper investigates the application of parameter weighting for linear prediction cepstral coefficients (LPCCs) to provide the robust representation of infant cry signals. Three classes of infant cry signals were considered such as normal cry signals, cry signals from deaf babies and babies with asphyxia. A Probabilistic Neural Network (PNN) is suggested to classify the infant cry signals into normal and pathological cries. PNN is trained with different spread factor or smoothing parameter to obtain better classification accuracy. The experimental results demonstrate that the suggested features and classification algorithms give very promising classification accuracy of above 98% and it expounds that the suggested method can be used to help medical professionals for diagnosing pathological status of an infant from cry signals.