Fadzil Ahmad

Intelligent medical disease diagnosis using improved hybrid genetic algorithm--multilayer perceptron network.

An improved genetic algorithm procedure is introduced in this work based on the theory of the most highly fit parents (both male and female) are most likely to produce healthiest offspring. It avoids the destruction of near optimal information and promotes further search around the potential region by encouraging the exchange of highly important information among the fittest solution. A novel crossover technique called Segmented Multi-chromosome Crossover is also introduced. It maintains the information contained in gene segments and allows offspring to inherit information from multiple parent chromosomes. The improved GA is applied for the automatic and simultaneous parameter optimization and feature selection of multi-layer perceptron network in medical disease diagnosis. Compared to the previous works, the average accuracy of the proposed algorithm is the best among all algorithms for diabetes and heart dataset, and the second best for cancer dataset.

A genetic algorithm-neural network approach for Mycobacterium tuberculosis detection in Ziehl-Neelsen stained tissue slide images

This paper describes a method using image processing and genetic algorithm-neural network (GA-NN) for automated Mycobacterium tuberculosis detection in tissues. The proposed method can be used to assist pathologists in tuberculosis (TB) diagnosis from tissue sections and replace the conventional manual screening process, which is time-consuming and labour-intensive. The approach consists of image segmentation, feature extraction and identification. It uses Ziehl-Neelsen stained tissue slides images which are acquired using a digital camera attached to a light microscope for diagnosis. To separate the tubercle bacilli from its background, moving k-mean clustering that uses C-Y colour information is applied. Then, seven Hus moment invariants are extracted as features to represent the bacilli. Finally, based on the input features, a GA-NN approach is used to classify into two classes: ‘true TB’ and ‘possible TB’. In this study, genetic algorithm (GA) is applied to select significant input features for neural network (NN). Experimental results demonstrated that the GA-NN approach able to produce better performance with fewer input features compared to the standard NN approach.

Performance comparison of gradient descent and Genetic Algorithm based Artificial Neural Networks training

One of the major issues concerning the Artificial Neural Networks (ANNs) design is a proper adjustment of the weights of the network. There have been a number of studies comparing the performance of evolutionary and gradient based ANNs learning. But the results of the studies, sometime conflicting to each other although the same and standard dataset development had been used. Motivated by this finding, the main objective of this paper is to make another comparison between the variations of gradient descent and Genetic Algorithm (GA) based ANNs training with special emphasize given on the developed algorithm and comparison methodology. Besides, the effect of the crossover operation on GA training is also being investigated. The comparison is done using cancer and diabetes benchmark dataset. The result shows that the overall classification error percentage of the family of GA is slightly better than those of gradient descent on cancer dataset. On the other hand, gradient descent is much better than GA on diabetes.

Improvement of Features Extraction Process and Classification of Cervical Cancer for the NeuralPap System

Abstract Cervical cancer has caused many deaths each year. Screening tests, such as Pap smear test used for the detection of the precancerous stage are able to avoid the occurrence of cervical cancer. However, the Pap smear test has several disadvantages such as less effective slides preparation and human error. Therefore, a computer-aided diagnosis system is introduced as a solution to the problem. One of the diagnostic systems that has been built is NeuralPap. However, the NeuralPap performance is limited by several constraints. This research proposed several new image processing algorithms to reduce these constraints. The Adaptive Fuzzy-k-Means (AFKM) clustering algorithm is proposed to replace the Moving k-Means (MKM) to segment Pap smear images into the nucleus, cytoplasm and background regions. Next, the feature extraction algorithm based on pseudo colouring called the Pseudo Colour Feature Extraction (PCFE) manual and Semi-Automatic PCFE are designed to replace the Region Growing Based Feature Extraction (RGBFE) which uses monochromatic images. This research is a step forward compared with the NeuralPap system by proposing the feature extraction algorithm for overlapping cells by combining the concept of colour space with Semi-Automatic PCFE algorithm. In addition, this research has also suggested the AFKM algorithm as a new centre positioning algorithm for the Radial Basis Function (RBF) and Hybrid RBF (HRBF) networks replacing the MKM algorithm. The entire proposed algorithm has been proven to produce better performance than the corresponding algorithm used in the NeuralPap. In addition, the combination of all algorithms has managed to increase the accuracy of the classification of cervical cancer to 76.35%, compared with 73.40% which is obtained from the previous NeuralPap system.

Intelligent Breast Cancer Diagnosis Using Hybrid GA-ANN

Breast cancer prevails as one of the infamous deathly diseases among women worldwide. Early detection and treatment of breast cancer can increase the survival rate of patients. Presently, the method of diagnosis depends on the human experiences. The method is time-consuming, subjected to human error and cause unnecessary burden to radiologists. This paper introduces an automatic breast cancer diagnosis technique using a genetic algorithm (GA) for simultaneous feature selection and parameter optimization of artificial neural networks (ANN). The performances of the proposed algorithm employing three different variations of the backpropagation technique for the fine tuning of the weight of ANN are compared. The algorithm is called the GAANN_XX where the XX refers to the back-propagation training variation used. The proposed algorithms called GAANN_RP produces the best and average, 99.43% and 98.29% correct classification respectively on the Wiscinson Breast Cancer Dataset.

Gel Electrophoresis Image Segmentation with Kapur Method Based on Particle Swarm Optimization

Gel electrophoresis (GE) is an important tool in genomic analysis. It is a process of DNA, RNA and protein molecules separation using electric field applied to a gel matrix. This paper describes the image processing techniques applied on GE image to segment the bands from their background. Numerous pre-processing steps are applied on the image prior to the segmentation technique for the purpose of removing noise in the image. Then multilevel thresholding using Otsu method based on Particle Swarm Optimization is applied. The experimental results show that the PSO-Otsu successfully segmented all the bands.

Optimization of FLC parameters for optimal control of FES-assisted elliptical stepping exercise using GA and PSO

This paper presents the parameter optimization of the fuzzy logic controller (FLC) for the Functional Electrical Stimulation (FES)-assisted elliptical stepping exercise. The FLC is used to control the cadence of the elliptical stepping exercise for smooth exercise movement. Genetic algorithm (GA) and particle swarm optimization (PSO) are used to optimize the parameters of the FLC. Both algorithms are implemented in Matlab and simulated with the dynamic model of the elliptical stepping exercise. In the performance analysis, the GA has faster convergence compared to the PSO where both converged at 40th and 51st iterations, respectively. The root mean square error (RMSE) for the GA and PSO are 7.873 rpm and 7.087 rpm respectively showing that the PSO has better performance in terms of the RMSE. Both techniques also have shown good performance in stepping cycle completion. The use of the GA and PSO had led towards more efficient FLC control for the FES-assisted elliptical stepping exercise.

Parameter Optimization of FES-Assisted Indoor Rowing Exercise Using MOGA

This paper describes the parameter optimization of fuzzy logic control (FLC) of FES-assisted indoor rowing exercise (FES-rowing) using multi objective genetic algorithm (MOGA). The indoor rowing exercise is introduced as a total body exercise for rehabilitation of function of lower extremities through the application of functional electrical stimulation (FES). FLC is used to control the knee and elbow trajectories for smooth rowing manoeuvre. MOGA is implemented in Matlab with a dynamic simulation model of indoor rowing exercise is developed using Visual Nastran (vN4D) software environment. MOGA is used to optimize the FES-rowing with two objective functions specified that are i) to minimize the mean squared error of knee angle trajectory and ii) to minimize the total electrical stimulation required by the muscles. In view of good results obtained, it is concluded that MOGA is able to obtain the optimal design of FLC for FES-rowing with two conflicting objectives.

Geopolymer Application in Soil: A Short Review

Geopolymer is a well-known material names by Davidovit’s since 1970’s. The other names of geopolymer is alkali-activated cement, geocement, alkali-bonded ceramic, inorganic polymer concrete, and hydroceramic. In a simple explanation, the termed ‘geopolymers’ comes when the inorganic polymeric material synthesized in a manner similar to thermosetting organic polymers. The development and contribution of geopolymer to the industries are moving stage by stage until today. Since a decades, performance of Geopolymer has been evaluated and tested by researchers in many field. The result published showed the unique bonding between aluminosilicate and alkali solution produce high compressive strength, low shrinkage, resistance toward acid, resistance to fire and etc. Advance research showed the application of Geopolymer in civil engineering works (including structures and geotechnical) also giving a good strength result. To that extend, this paper try to review performances of geopolymer application in geotechnical fields.

Muscle Extension Model for FES-Assisted Knee Swinging Ergometer for Stroke Patient

This work is focused on the development of physiological and mathematical model of muscle extension induced by FES-assisted knee swinging ergometer. A muscle model composed of muscle activation, muscle contraction and body segmental dynamics is developed to perform full knee extension. The developed muscle model is then incorporated with a humanoid model and a knee swinging ergometer model to perform simulation of FES-assisted knee swinging exercise. The developed humanoid model possesses the characteristics of stroke patient while the knee swinging ergometer was developed with the ability of utilizing the non-paretic knee in assisting the swinging of the paretic knee. The developed muscle model is evaluated under the dynamic simulation of FES-assisted knee swinging ergometer. The active torque and knee trajectories are recorded and the results shows that the design of knee swinging ergometer had successfully reduced the required stimulation pulse width in performing full knee extension. It is also suggested that the muscle model developed can be used for future analysis and control development of FES-assisted knee swinging ergometer for stroke patient.