Wan Khairunizam Wan Ahmad
Thumbnail Image
Preferred name
Wan Khairunizam Wan Ahmad
Official Name
Wan Khairunizam, Wan Ahmad
Alternative Name
Wan, Khairunizam
Ahmad, Wan Khairunizam Wan
Khairunizam, W. A. N.
Main Affiliation
Scopus Author ID
57200576499
Researcher ID
E-6072-2011

Research Output

Filters

78
Reset filters

Settings
Now showing 1 - 10 of 78
  • Publication
    Performance Comparison of Machine Learning Algorithms for Classification of Chronic Kidney Disease (CKD)
    ( 2020-06-17)
    Azian Azamimi Abdullah
    ;
    Hafidz S.A.
    ;
    Wan Khairunizam Wan Ahmad
    Kidney is one of the vital organs in a human body while ironically, chronic kidney disease (CKD) is one of the main causes of death in the world. Due to the low rate of loss of kidney function, the disease is often overlooked until it is in a really bad condition. Dysfunctional kidney may lead to accumulation of wastes in blood which would affect several other systems and functions of the body such as blood pressure, red blood cell production, vitamin D and bone health. Machine learning algorithms can help in classifying the patients who have CKD or not. Even though several studies have been made to classify CKD on patients using machine-learning tool, not many researchers perform pre-processing and feature selection technique to obtain quality and dependable result. Machine learning used with feature selection techniques are shown to have better and more dependable result. In this study, feature selection methods such as Random Forest feature selection, forward selection, forward exhaustive selection, backward selection and backward exhaustive selection were identified and evaluated. Then, machine learning classifiers such as Random Forest, Linear and Radial SVM, Naïve Bayes and Logistic Regression were implemented. Lastly, the performance of each machine-learning model was evaluated in terms of accuracy, sensitivity, specificity and AUC score. The results showed that Random Forest classifier with Random Forest feature selection is the most suitable machine learning model for classification of CKD as it has the highest accuracy, sensitivity, specificity and AUC with 98.825%, 98.04%, 100% and 98.9% respectively which outperformed other classifiers.
  • Publication
    A review of the application and effectiveness of heat storage system using phase change materials in the built environment
    ( 2021-05-03)
    Ibrahim Z.
    ;
    Newby S.
    ;
    Hassani V.
    ;
    Ya'akub S.R.
    ;
    Shahriman Abu Bakar
    ;
    Zuradzman Mohamad Razlan
    ;
    Wan Khairunizam Wan Ahmad
    Global warming is the most significant threat that civilization faced within the 21st century. Buildings, which account for 40% of global consumption of energy and greenhouse gas emissions, play a key role in global warming. It is estimated that their destructive impact will grow by 1.8 percent per year by 2050, indicating that future energy consumption and emissions will be more critical than they are today. Therefore, the use of a latent heat storage system using phase change materials (PCM) is one of the effective ways of storing thermal energy and has the advantages of high-energy storage density and the isothermal nature of the storage process. PCM has been widely used in latent heat thermal storage systems for heat pumps, solar engineering, and spacecraft thermal control applications. Thermal energy conservation by latent heat is an ideal way to increase the thermal inertia of building envelopes, which would minimize temperature fluctuations, contributing to increased occupants' thermal comfort. For this reason, high-density PCM can be used effectively. This paper reviews recent studies of the application and effectiveness of using PCM in the built environment.
  • Publication
    Tensile characterizations of oil palm empty fruit bunch (Opefb) fibres reinforced composites in various epoxy/fibre fractions
    ( 2022-10-15)
    Mohd Khairul Faizi Abd Rahman
    ;
    Shahriman Abu Bakar
    ;
    Mohd Shukry Abdul Majid
    ;
    Tamrin S.B.M.
    ;
    Israr H.A.
    ;
    Anas Abdul Rahman
    ;
    Guan N.Y.
    ;
    Zuradzman Mohamad Razlan
    ;
    Kamis N.A.
    ;
    Wan Khairunizam Wan Ahmad
    Oil palm empty fruit bunch (OPEFB) single fibers and reinforced composites were comprehensively characterized through tensile tests to assess their performance as potential reinforcing materials in polymer composites. The performances of OPEFB single fibers and reinforced composites with untreated and treated fibers conditions were compared. The fibers were variously treated with 3% sodium hydroxide, 2% silane, 3% sodium hydroxide mixed with 2% silane, and 3% sodium hydroxide prior to 2% silane for 2 hours soaking time. The highest toughness of the single fibers test was then selected to proceed with composites fabrication. The OPEFB composites were fabricated in 90:10, 80:20, 70:30, and 60:40 epoxy-fibre fractions. The result shows that the selected treated fiber composite exhibits better performance. The selected treated fiber composite increased the highest ultimate tensile strength by 145.3% for the 90:10 fraction. The highest Young’s Modulus was increased by about 166.7% for 70:30 fraction. Next, the highest toughness was increased by 389.5% for the 30:70 fraction. The treated fibers provided a better interlocking mechanism between the matrix and fibers in reinforced composites, thus improving their interfacial bonding.
  • Publication
    Niblack algorithm modification using maximum-minimum (Max-min) intensity approaches on low contrast document images
    ( 2020-01-01)
    Wan Azani Wan Mustafa
    ;
    Wan Khairunizam Wan Ahmad
    ;
    Mat Yusoff A.S.
    ;
    Syed Zulkarnain Syed Idrus
    ;
    Mohamad Nur Khairul Hafizi Rohani
    In recent decades, detection or segmentation has been one of the major interesting research subjects due to the analysis of the information. However, most of the historical document has degraded and low contrast problem. Recently, many binarization methods were proposed in order to segment the text region from the background region in the low-quality image. In this paper, an improved binarization method was inspired by Niblack method was presented. The modification focuses to find the optimum threshold value by using the Maximum-Minimum intensity technique. The main target is to reduce the unwanted detection image and increase the resultant performance compared to the original Niblack method. The proposed method was applied to the document images from H-DIBCO 2012 and H-DIBCO 2014 dataset. The results of the numerical simulation indicate that the target was achieved by the F-Measure by F-measure (58.706), PSNR (10.778) and Accuracy (86.876). This finding will give a new benchmark to other researchers to propose an advance binarization method.
  • Publication
    Blood vessel detection monitoring system and mobile notification for diabetic retinopathy diagnosis
    ( 2020-01-01)
    Mahmud A.S.
    ;
    Wan Azani Wan Mustafa
    ;
    Mohd Aminudin Jamlos
    ;
    Syed Zulkarnain Syed Idrus
    ;
    Wan Khairunizam Wan Ahmad
    ;
    Mohd Al-Hafiz Mohd Nawi
    Disease diagnosis based on retinal image analysis is very popular in order to detect a few critical diseases such as diabetic retinopathy, high blood pressure, cancer and glaucoma. The important part of the retinal is a blood vessel. Besides, the blood vessel study plays an important part in different medical areas such as ophthalmology, oncology, and neurosurgery. The significance of the vessel analysis was helped by the continuous overview in clinical studies of new medical technologies intended for improving the visualization of vessels. In this paper, a new blood vessel detection based on a combination of Kirsch’s templates and Fuzzy C-Means (FCM) was proposed. The main objective of this study is to improve the detection result of FCM and achieved more effective performance compared to the Kirsch’s templates result. The proposed method experimented on 20 images is utilized namely from Digital Retina Images for Vessel Extraction (DRIVE) dataset. The resulting images are compared with the benchmark images based on a few image quality assessment (IQA) such as accuracy, sensitivity and specificity. The total average of accuracy is 92.64%, while sensitivity and specificity obtained was 95.73% and 60.45% respectively. The three parameters of the IQA will then be displayed in a column on the GUI. The second part of the system is for the mobile notification system to send SMS to a mobile phone. In order for the user to obtain the image analysis results, there must be a notification system on the mobile phone. By using the GSM module integrated with Arduino Uno, notification regarding image analysis will be sent to the mobile phone.
  • Publication
    No-reference quality assessment for imagebased assessment of economically important tropical woods
    ( 2020-05-01)
    Rajagopal H.
    ;
    Mokhtar N.
    ;
    Izam T.F.T.M.N.
    ;
    Wan Khairunizam Wan Ahmad
    Image Quality Assessment (IQA) is essential for the accuracy of systems for automatic recognition of tree species for wood samples. In this study, a No-Reference IQA (NR-IQA), wood NR-IQA (WNR-IQA) metric was proposed to assess the quality of wood images. Support Vector Regression (SVR) was trained using Generalized Gaussian Distribution (GGD) and Asymmetric Generalized Gaussian Distribution (AGGD) features, which were measured for wood images. Meanwhile, the Mean Opinion Score (MOS) was obtained from the subjective evaluation. This was followed by a comparison between the proposed IQA metric, WNR-IQA, and three established NR-IQA metrics, namely Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE), deepIQA, Deep Bilinear Convolutional Neural Networks (DB-CNN), and five Full Reference-IQA (FR-IQA) metrics known as MSSIM, SSIM, FSIM, IWSSIM, and GMSD. The proposed WNR-IQA metric, BRISQUE, deepIQA, DB-CNN, and FR-IQAs were then compared with MOS values to evaluate the performance of the automatic IQA metrics. As a result, the WNR-IQA metric exhibited a higher performance compared to BRISQUE, deepIQA, DB-CNN, and FR-IQA metrics. Highest quality images may not be routinely available due to logistic factors, such as dust, poor illumination, and hot environment present in the timber industry. Moreover, motion blur could occur due to the relative motion between the camera and the wood slice. Therefore, the advantage of WNRIQA could be seen from its independency from a "perfect" reference image for the image quality evaluation.
  • Publication
    Effect of different filtering techniques on medical and document image
    ( 2021-01-01)
    Wan Azani Wan Mustafa
    ;
    Sam S.
    ;
    Mohd Aminudin Jamlos
    ;
    Wan Khairunizam Wan Ahmad
    Image enhancement is very important stages used in image processing. A normal image enhancement process is using the filtering technique. Filtering helps the problems of the image display and can improvise the quality of the image. The problems that always happened in the image is illumination, noise and under-light images. In addition, these problems also caused a few troubles for image recognition for the daily life of certain people for their work. The objective of this study is to explore and compare a few starts of art filtering techniques based on the mathematical algorithm of the filters and then identifying the best method of the filters. There were a few methods that were selected in this project such as a high pass filter, low pass filter, high boost filter and others. All the selected filter experimented on the medical images and document images. The resulting images were evaluated using the Image Quality Assessments (IQA) which is a global contrast factor (GCF) and signal to noise ratio (SNR). Based on the numerical result, homomorphic low pas filter (HLF) provides a better performance among the other filters in terms of GCF (2.066) and SNR (8.907) value of the selected images.
  • Publication
    Time Domain Analysis for Emotional EEG Signals of Stroke Patient and Normal Subject
    ( 2023-01-01)
    Vincen E.
    ;
    Wan Khairunizam Wan Ahmad
    ;
    Yean C.W.
    ;
    Wan Azani Wan Mustafa
    This paper aims to analyze the emotional Electroencephalogram (EEG) signals of different time windows. The time window of the signals is one of the variables that affect the efficiency of the EEG signal analysis. In this research, a total of 30 subjects are analyzed from three different groups namely 10 left brain damage (LBD), 10 right brain damage (RBD), and 10 normal control (NC) for six different emotional states. The 14-Channel Wireless Emotiv EPOC device with a sampling frequency of 128 Hz is used to extract EEG signal from the subjects. The 6th Order Butterworth Bandpass filter is used to extract the EEG signals with the frequency band of 8-49 Hz, which are alpha to gamma waves. The EEG signals are segmented in 2s, 4s, 6s, and 8s time windows for all frequency bands. In addition, the K-Nearest Neighbor (KNN) and Probabilistic Neural Network (PNN) classifiers are used to classify the six emotions in LBD, RBD and NC. The beta and gamma bands are the best performing EEG frequency band for emotion classification. In the investigation, 6s time windows have the highest classification accuracy for KNN with 81.90% and 8s time window for PNN classifier with 82.15%.
  • Publication
    An Aggressiveness Level Analysis Based on Buss Perry Questionnaire (BPQ) and Brain Signal (EEG)
    ( 2021-12-01)
    Munian L.
    ;
    Wan Khairunizam Wan Ahmad
    ;
    Xu T.K.
    ;
    Wan Azani Wan Mustafa
    ;
    Rahim M.A.
    Aggression is the most important human aspects that make daily things possible for individuals, to succeed and have a better level of behaviour. Aggression is feelings of anger or antipathy resulting in hostile or violent behaviour. The importance of aggression is to increase an individual dominance of the subject in their social environment. Traditionally, the subject's aggression is usually measured by using a survey through Buss-Perry Questionnaire (BPQ). Considering the variability of the aggressiveness level, this study proposes investigation of aggression by using BPQ and Electroencephalography (EEG) to evaluate the aggressiveness level of the subjects. The results of the BPQ are analysed based on the final score that are responded by the subjects. In EEG experiment, the evaluation of subject's aggressiveness while playing a smart phone game “Subway Surfers”, a basic method has been employed, namely correlation coefficient method. The EEG signals are recorded while the subject playing the game.The number of subjects involves in the experiment is 9 and they are the UniMAP's male students at the age of 21-25 years old. In the analysis, the induced aggression is compared between BPQ with Net Aggressiveness Index (NAI), which is obtained from brain signals (EEG). The BPQ obtains the subjects #2 and subject #8 are the highest Buss-Perry Aggresinenes Index (BPAI) scores, which are 0.32244 and 0.32223 respectively. Meanwhile in EEG analysis the subject #8 only achieves the highest score of 0.34713. From the results of the investigation, it could be concludedthat the use of EEG to identify the aggressiveness level will overcome the disadvantage of the conventional methods.
  • Publication
    Progress Monitoring in Upper Limb Stroke Rehabilitation by Using Muscle Activation and Hand Speed
    ( 2020-06-17)
    Lee H.L.
    ;
    Wan Khairunizam Wan Ahmad
    ;
    Cahyadi B.N.
    ;
    Wan Azani Wan Mustafa
    ;
    Syed Zulkarnain Syed Idrus
    Nowadays, Virtual Reality (VR) technology is commonly used in the rehabilitation to increase the motivation of the stroke patients do the exercises, however, very few researches to monitor the rehabilitation progress has reported. VR based rehabilitation is interesting because could motivate stroke patients in their long-term rehabilitation process. This research is to evaluate the progress monitoring of the subject after conducting three sessions of rehabilitation exercise. Five male and female healthy subjects were selected to do the rehabilitation game. Three VR games are designed for the subject to perform three different movement sequences. The selected upper limb characteristics are muscle activation and hand speed. An Electromyography (EMG) is used to measure the muscle activation through an electrical activity of the muscle, while a Kinect sensor is used to measure the hand speed. The experimental results show that the proposed upper limb characteristics are able to be used for monitoring progress in the rehabilitation.