Wan Azani Wan Mustafa
Thumbnail Image
Preferred name
Wan Azani Wan Mustafa
Official Name
Wan Azani, Wan Mustafa
Alternative Name
Mustafa, W.
Azani Mustafa, Wan
Mustaffa, Wan Azani
Wan Mustafa, Wan Azani
Main Affiliation
Scopus Author ID
57219421621
Researcher ID
J-4603-2014

Research Output

Filters

2
4
Reset filters

Settings
Now showing 1 - 4 of 4
Thumbnail Image
Publication

Heart Arrhythmia Classification Using Deep Learning: A Comparative Study

2023 , Radi Omar , Alslatie Mohammad , Wan Azani Wan Mustafa , Alquran Hiam , Badarneh Alaa , Mohammed F.F. , Ahmed Alkhayyat

Heart arrhythmia is an irregular heartbeat that causes heart problems. It can be classified by their seriousness into serious and non-serious arrhythmia. Mainly to diagnose heart arrhythmias, we use Electrocardiogram (ECG). In this paper, the authors compared three different models of classifiers: Convolutional Neural Network, Dense Neural Network and Long Short-Term Memory to classify cardiac arrhythmia into two types normal and abnormal, using the MIT-BIH database. The results show that CNN and DNN have the best result of the models with 99% accuracy while LSTM shows 60 accuracy percent.

View
No Thumbnail Available
Publication

Nucleus Detection Using Deep Learning Approach on Pap Smear Images

2023 , Alquran Hiam , Wan Azani Wan Mustafa , Mohammed F.F. , Alkhayyat Ahmed

Cervical cancer is caused by the abnormal growth of female cervix cells. It is one of the most familiar factors for women's death worldwide. Therefore., early detection of cervical cancer leads to a reduced mortality rate and increased chance of being alive. The Papanicolaou is a common method for screening and identifying the cancerous cells in a woman's cervix. The resultant pap smear images may help the physician diagnose the cervix cells. The crucial part of the cell is the nucleus. Therefore., auto-detection of the nucleus is the core point in this paper. A deep learning algorithm is employed to segment the nucleus in pap smear images. Two network structures., known as ResNet18 and ResNet50., are exploited to detect the nucleus part in the cell. The results are compared with ground truth and between the two structures. Both networks., ResNet18 and ResNet50., perform almost the same., with test accuracy reaching 92%. This work distinguishes it from other work in simplicity., fast., and accuracy. Therefore., it can be recommended to be used in clinical units and rural countries which suffer from the lack of specialist physician.

View
No Thumbnail Available
Publication

Stages Classification on Cervical Cell Images: A Comparative Study

2023 , Wan Azani Wan Mustafa , Mohamad Irfan Noor , Alquran Hiam , Miharaini Md Ghani , Hafizul Fahri Hanafi , Noor Hidayah Che Lah , Mundher Adnan M. , Hameed Abdul Hussein Abbas

The cancer of the cervix is called cervical cancer. An element of a woman's womb is the cervix. Among other diseases that affect women, it came in at number four on the list. According to the World Health Organization's cancer report, there are currently roughly 10 million new cases of cancer recorded year, and by 2020, that number will have doubled to 20 million. With the right screening and awareness campaign, this number can be cut in half. A quarter of cancers are said to be brought on by infections, including hepatitis B, which is connected to liver cancer, and the human papillomavirus, which is connected to cervix cancer. Deep learning techniques have been successfully applied to a wide range of image classification tasks, and have the potential to be highly effective for cervical cell image classification as well. In this project, we propose to use a deep learning-based approach to classify cervical cell images into different categories, such as normal cells, abnormal cells, or cancerous cells. To achieve this goal, we will first pre-process the images to prepare them for analysis, and then extract relevant features. These features will be used to train a deep learning model, which will be fine-tuned and optimized for the specific task of cervical cell classification. In this project, transfer learning method will be by using pre-trained classifier such as ResNet-50, GoogLeNet and EfficientNet-b0. We will evaluate the performance of the model using metrics such as accuracy and compare our results to those obtained using traditional machine learning approaches. From this project, the highest accuracy achieved are 51.49%. The goal to develop a pre-trained classifier transfer learning can be used to accurately and reliably classify cervical cell images in a clinical setting are achieved.

View
Thumbnail Image
Publication

A New Weighted Deep Learning Feature Using Particle Swarm and Ant Lion Optimization for Cervical Cancer Diagnosis on Pap Smear Images

2023 , Alsalatie Mohammed , Alquran Hiam , Wan Azani Wan Mustafa , Zyout Ala’a , Alqudah Ali Mohammad , Kaifi Reham , Qudsieh Suhair

One of the most widespread health issues affecting women is cervical cancer. Early detection of cervical cancer through improved screening strategies will reduce cervical cancer-related morbidity and mortality rates worldwide. Using a Pap smear image is a novel method for detecting cervical cancer. Previous studies have focused on whole Pap smear images or extracted nuclei to detect cervical cancer. In this paper, we compared three scenarios of the entire cell, cytoplasm region, or nucleus region only into seven classes of cervical cancer. After applying image augmentation to solve imbalanced data problems, automated features are extracted using three pre-trained convolutional neural networks: AlexNet, DarkNet 19, and NasNet. There are twenty-one features as a result of these scenario combinations. The most important features are split into ten features by the principal component analysis, which reduces the dimensionality. This study employs feature weighting to create an efficient computer-aided cervical cancer diagnosis system. The optimization procedure uses the new evolutionary algorithms known as Ant lion optimization (ALO) and particle swarm optimization (PSO). Finally, two types of machine learning algorithms, support vector machine classifier, and random forest classifier, have been used in this paper to perform classification jobs. With a 99.5% accuracy rate for seven classes using the PSO algorithm, the SVM classifier outperformed the RF, which had a 98.9% accuracy rate in the same region. Our outcome is superior to other studies that used seven classes because of this focus on the tissues rather than just the nucleus. This method will aid physicians in diagnosing precancerous and early-stage cervical cancer by depending on the tissues, rather than on the nucleus. The result can be enhanced using a significant amount of data.

View