Mogeeb A. A. Mosleh

A preliminary study on automated freshwater algae recognition and classification system

BackgroundFreshwater algae can be used as indicators to monitor freshwater ecosystem condition. Algae react quickly and predictably to a broad range of pollutants. Thus they provide early signals of worsening environment. This study was carried out to develop a computer-based image processing technique to automatically detect, recognize, and identify algae genera from the divisions Bacillariophyta, Chlorophyta and Cyanobacteria in Putrajaya Lake. Literature shows that most automated analyses and identification of algae images were limited to only one type of algae. Automated identification system for tropical freshwater algae is even non-existent and this study is partly to fill this gap.ResultsThe development of the automated freshwater algae detection system involved image preprocessing, segmentation, feature extraction and classification by using Artificial neural networks (ANN). Image preprocessing was used to improve contrast and remove noise. Image segmentation using canny edge detection algorithm was then carried out on binary image to detect the algae and its boundaries. Feature extraction process was applied to extract specific feature parameters from algae image to obtain some shape and texture features of selected algae such as shape, area, perimeter, minor and major axes, and finally Fourier spectrum with principal component analysis (PCA) was applied to extract some of algae feature texture. Artificial neural network (ANN) is used to classify algae images based on the extracted features. Feed-forward multilayer perceptron network was initialized with back propagation error algorithm, and trained with extracted database features of algae image samples. Systems accuracy rate was obtained by comparing the results between the manual and automated classifying methods. The developed system was able to identify 93 images of selected freshwater algae genera from a total of 100 tested images which yielded accuracy rate of 93%.ConclusionsThis study demonstrated application of automated algae recognition of five genera of freshwater algae. The result indicated that MLP is sufficient, and can be used for classification of freshwater algae. However for future studies, application of support vector machine (SVM) and radial basis function (RBF) should be considered for better classifying as the number of algae species studied increases.

Challenges of Digital Note Taking

There are world efforts to make technology act with education field for better learning achievements. Technology tries to replace the traditional learning environments, media, and activities into digital age. However, slow progress has been achieved to transfer the note taking activities into digital era. In this study, we explored current note taking tools which developed to bridge the gap between paper-based and technology-based notes. We tried to identify key specific problems and challenges that prevent note taking from existing in the digital age. This study is providing extensive investigation with systematic analysis about the impacts of current note taking tools in learning to identify constrains and limitations of typical note taking systems. Unfortunately, we agreed with similar previous studies that current tools are still inadequate and inefficient to be used for replacing the traditional note taking due to several issues. We found that developing a successful note taking applications is challenges because of four main issues, complexity, technology learning dilemma, integrity, and inefficiency issues. This study discusses the main implications to shape the future of digital notes.

Random Forest and Self Organizing Maps Application for Analysis of Pediatric Fracture Healing Time of the Lower Limb

Abstract In this study, we examined the lower limb fracture healing time in children using random forest (RF) and Self Organizing feature Maps (SOM) methods. The study sample was obtained from the pediatric orthopedic unit in University Malaya Medical Centre. Radiographs of long bones of lower limb fractures involving the femur, tibia and fibula from children ages 0–12 years, with ages recorded from the date and time of initial injury. Inputs assessment extracted from radiographic images included the following features: type of fracture, angulation of the fracture, contact area percentage of the fracture, age, gender, bone type, type of fracture, and number of bone involved. RF is initially used to rank the most important variables that effecting bone healing time. Then, SOM was applied for analysis of the relationship between the selected variables with fracture healing time. Due to the limitation of available dataset, leave one out technique was applied to enhance the reliability of RF. Results showed that age and contact area percentage of fracture were identified as the most important variables in explaining the fracture healing time. RF and SOM applications have not been reported in the field of pediatric orthopedics. We concluded that the combination of RF and SOM techniques can be used to assist in the analysis of pediatric fracture healing time efficiently.

A Primary Study on Application of Artificial Neural Network in Classification of Pediatric Fracture Healing Time of the Lower Limb

In this study we examined the lower limb fracture in children and classified the healing time using supervised and unsupervised artificial neural network (ANN). Radiographs of long bones from 2009 to 2011 of lower limb fractures involving the femur, tibia and fibula from children ages 0 to 13 years, with ages recorded from the date and time of initial injury was obtained from the pediatric orthopedic unit in University Malaya Medical Centre. ANNs was developed using the following input: type of fracture, angulation of the fracture, displacement of the fracture, contact area of the fracture and age. Fracture healing time was classified into two classes that is less than 12 weeks which represent normal healing time in lower limb fractures and more than 12 weeks which could indicate a delayed union. This research was designed to evaluate the classification accuracy of two ANN methods (SOM, and MLP) on pediatric fracture healing. Standard feed-forward, back-propagation neural network with three layers was used in this study. The less sensitive variables were eliminated using the backward elimination method, and the ANN network was retrained again with minimum variables. Accuracy rate, area under the curve (AUC), and root mean square errors (RMSE) are the main criteria used to evaluate the ANN model results. We found that the best ANN model results was obtained when all input variables were used with overall accuracy percentage of 80%, with RMSE value of 0.34, and AUC value of 0.8. We concluded here that the ANN model in this study can be used to classify pediatric fracture healing time, however extra efforts are required to adapt the ANN model well by using its full potential features to improve the ANN performance especially in the pediatric orthopedic application.

Automated plant identification using artificial neural network and support vector machine

ABSTRACT Ficus is one of the largest genera in plant kingdom reaching to about 1000 species worldwide. While taxonomic keys are available for identifying most species of Ficus, it is very difficult and time consuming for interpretation by a nonprofessional thus requires highly trained taxonomists. The purpose of the current study is to develop an efficient baseline automated system, using image processing with pattern recognition approach, to identify three species of Ficus, which have similar leaf morphology. Leaf images from three different Ficus species namely F. benjamina, F. pellucidopunctata and F. sumatrana were selected. A total of 54 leaf image samples were used in this study. Three main steps that are image pre-processing, feature extraction and recognition were carried out to develop the proposed system. Artificial neural network (ANN) and support vector machine (SVM) were the implemented recognition models. Evaluation results showed the ability of the proposed system to recognize leaf images with an accuracy of 83.3%. However, the ANN model performed slightly better using the AUC evaluation criteria. The system developed in the current study is able to classify the selected Ficus species with acceptable accuracy.

Reviewing and Classification of Software Model Checking Tools

In this study, we provide historical accounts with an overview of essential research on model-checking development tools. This study has two main objectives; first, it is intended to investigate whether model checking still an active area; second, to classify existing model-checking tools by providing an illustration of each dimension scope, an analysis of similarities and differences among them, and a prediction of the future direction of typical model-checking tools. We found that existing model-checking tools show significant effects in automated system testing and verification. We also found that system testing and verification are still active areas of research. Current model-checking tools work efficiently on limited environment, and a lot of work need to perform for verifying the functional and nonfunctional attributes of complex systems. Despite the limitations of existing model-checking tools, universal model-checking tools can probably be developed if a good framework is established to fulfill the requirements of fully automated tools.

Ecological data prediction and visualization system

Temporal patterns in ecological data can be visualized and communicated effectively through graphical means. The aim of this study was to develop a data prediction and visualization system based on historical data and thematic map technology to visualize forecast temporal ecological changes. The visualization system consists of prediction and data visualization modules. The prediction module is developed using a hybrid evolutionary algorithm (HEA) to classify and predict noisy ecological data. The visualization module is developed using Dotnet Framework 2.0 to implement thematic cartography for volume visualization. The visualization system is evaluated by its capability in representing the output data on a map, and by predicting the abundance of Chlorophyta based on other water quality parameters. Rules for predicting Chlorophyta abundance had a success rate of almost 90%. The integration of computational data mining using HEA and visualization using thematic maps promises practical solutions and better techniques for forecasting temporal ecological changes, especially when data sets have complex relationships without clear distinction between various variables.

An image processing system for cephalometric analysis and measurements

Computerized cephalometric information enhance clinical and research studies by making the information accessible, consistent, and statistically valid in comparative studies. Image processing technique was used widely to solve many problems in medical images. This study has developed a new system to computerize the manual process of cephalometric. The system has contained three modules to perform cephalometric analysis and measurements. The filtering module was developed to enhance the contrast of x-ray images; the locating landmark module was developed to identify the interest points in x-ray images manually; and the measurement module was developed to perform angular and linear measurements of cephalometric. The filters make x-ray image clearer that landmark points can be easily identified. The developed system has reduced the processing time for obtaining results of cephalometric measurements for more than 10 times of the manual methods. . The results showed a great accuracy between manual and automatic methods using this system.