Zeinab A. Mustafa

Sudan University of Science and Technology

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Banazier A. Ibraheem

Sudan University of Science and Technology

Banazier A. Abrahim

Sudan University of Science and Technology

Yasser M. Kadah

Cairo University

Ethar A. Awad

Sudan University of Science and Technology

Inas A. Yassine

Cairo University

Nourhan Zayed

Electronics Research Institute

Sarah R. Omer

Sudan University of Science and Technology

Rania E. Elmahdi

Sudan University of Science and Technology

Sahar Haj Ali A. Mohammed

AlMughtaribeen University

Azza M. Bin Aof

Sudan University of Science and Technology

Cooperative Institutions

Sudan University of Science and Technology

University of Medical Sciences and Technology

AlMughtaribeen University

University of Gezira

University of Khartoum

King Abdulaziz University

Cairo University

Federal Ministry of Health

Al-Neelain University

University of KwaZulu-Natal

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Noise Reduction for Magnetic Resonance Imaging by Using Edge Detection and Hybrid Mean Lee Filter Techniques

Journal of Clinical Engineering (2022)

Zeinab A. Mustafa Banazier A. Ibraheem Kawther M. GissmAllah Rania E. Elmahdi Akram I. Omara

In this article, spatial-domain filtering algorithms were developed to suppress additive noise in magnetic resonance (MR) imaging. It is difficult to suppress MR image noise because it corrupts almost all pixels in an image. The purpose of noise reduction is to curb the noise with high efficiency while keeping the edges and other detailed features as much as possible. The present article focused on developing quite efficient noise reduction by using an edge detection technique and hybrid mean Lee filters to suppress MR image noise quite effectively in spatial domain without yielding much distortion and blurring. The performances of the developed filter were compared with the existing filters in terms of universal quality index, method noise, and execution time. Among all existing filters, the edge detection technique and hybrid mean Lee filter was found to be best for suppressing MR image noise.

Salt-and-pepper noise

Distortion (music)

Image noise

Dark-frame subtraction

Value noise

10.1097/jce.0000000000000559

Cite

Citations (2)

Fast phase aberration correction in ultrasound imaging using fat layer model

Banazier A. Abrahim Zeinab A. Mustafa Yasser M. Kadah

Phase aberration is one of the most important factors that limit improvement to lateral resolution of ultrasound imaging system. In this paper we propose a computationally efficient method to correct the phase aberration problem arises from the subcutaneous fat layer. The method is based on the determination of thickness of the fat layer and the velocity in it to calculate the focusing delay perfectly. The thickness and velocity can be determined manually by the user through a qualitative assessment or automatically using a quantitative measure as an objective function. Minimizing the value of the entropy was selected as the cost function. The effect of the fat layer thickness and velocity were simulated as a time delays added to the radio frequency (RF) data. Experimental studies addressing that the entropy can be used to accurately determine the thickness and velocity of the fat layer depending on the selected region of interest (ROI). Images of a six pins phantom were reconstructed by two method in frequency domain by Fourier transform and in time domain, and different images were reconstructed using different aperture size . We drive the evaluation of results simulation in image reconstructed by two method and evaluation phase aberration correction when we used best thickness only and when we used optimum velocity also we do it by execution time and accuracy of image using entropy cost function.

Point spread function

10.1109/icces.2007.4447048

Cite

Citations (1)

Breast Tumors Classification Using Adaptive Neuro-Fuzzy Inference System

Journal of Clinical Engineering (2017)

Sahar Haj Ali A. Mohammed Zeinab A. Mustafa

Breast cancer is one of the world's leading causes of cancer-related deaths and ranks second in the cancer fact sheets. In Sudan, the increasing incidence, detection at late stages, and early onset of the disease make early detection and diagnosis of breast cancer an overbearing task. The objective of this study was to create a computer interfacing system for the localization, detection, and classification of breast masses using adaptive neuro-fuzzy inference system (ANFIS). The ANFIS classifier was used to detect the breast cancer when 5 features defining breast cancer indications were used as inputs. The proposed ANFIS model combined the neural network adaptive capabilities and the fuzzy-logic qualitative approach. Results demonstrated that the proposed methodologies have high potential in enhancing breast images and localizing, detecting, and classifying the breast tumor. The system was able to achieve an accuracy of 94.4% sensitivity, 100% specificity, 97.1% positive predictive value, 100% negative predictive value, an Az value of .972, and an overall classification accuracy of 98%.

Fuzzy inference system

Interfacing

Inference system

10.1097/jce.0000000000000205

Cite

Citations (1)

Deep Learning Algorithms for Studying the Impact of Tumor Suppressor Gene Mutations on Breast Cancer

Journal of Clinical Engineering (2024)

Samah Gaysar Zeinab A. Mustafa A. Ould Zein

Breast cancer is the most common site of cancer causing death in women around the world. It is the most frequently diagnosed malignancy in women, and mutations in the tumor suppressor p53 are commonly detected in the most aggressive subtypes. Rapid advances in biological research over recent years have significantly enriched biological and medical data resources. Deep learning–based techniques have been successfully utilized to process data in this field, and they have exhibited state-of-the-art performances even on high-dimensional, nonstructural, and black-box biological data. In this article, we propose a deep learning approach by using recurrent neural networks to evaluate and assess the contribution of genetic mutations in the TP53 gene in the breast cancer. Moreover, preprocessing of the breast dataset (the genetic dataset used comprises TP53 gene sequences, for normal and breast cancer cases; 100 sequences of each class, obtained from NCBI, Ensembl, IGSR, and TCGA) was done by machine learning algorithms such as k-nearest neighbors and principal component analysis and artificial neural networks. The experimental results show that under a different dataset, the mutation on TP53 appears in about 80% of this dataset; accuracy achieved by the recurrent neural network model was 92%, and the precision was 91%. Finally, to enhance the performance and applicability of the model, it is recommended to focus on preprocessing stage and use different and cross-section modules.

10.1097/jce.0000000000000681

Cite

Citations (0)

An Innovative Leukemia Detection System using Blood Samples via a Microscopic Accessory

International Journal of Engineering and Manufacturing (2023)

Azza M. Bin Aof Ethar A. Awad Sarah R. Omer Banazier A. Ibraheem Zeinab A. Mustafa

Leukemic patients are in a rapid increase.Hence, the use of microscopic images of blood samples through visual inspection to identify blood disorders has increased, opening the door for computerized techniques for detecting leukemia.This project applies computer vision techniques to increase the accuracy and speed of detection from peripheral blood.It also enhances visualization by providing an appropriate supplement to traditional microscopy.A microcomputer (Raspberry Pi) was well programmed in Python for analyzing images with the help of a Raspberry Pi camera and a touch screen as an alternative to the eyepiece.To achieve diversity and seek for more accuracy, image datasets for this project were obtained from various resources.These datasets were then analyzed through image processing techniques to detect leukemia cells.This detection process involves resizing cells to a standard size, noise removal by linear scaling filter, global-local contrast enhancement, segmentation of white blood cells (WBCs) using marker-controlled watershed algorithm, overlapping detection and separation using watershed and k-means clustering algorithms, and extraction with selection of the most relevant features from cells.These features were then imported into the Support Vector Machine (SVM) model which resulted in an accuracy of 93.2773%.A standalone desktop application with a suitable graphical user interface (GUI) was implemented.It was then uploaded into the Raspberry Pi, some code lines were rewritten for dealing with the camera, the hardware was designed and implemented, and then formal experiments were conducted resulting in the detection of leukemia in 5 samples out of 6.This implies that precise detection can be implemented with different data taken in various imaging conditions.

Raspberry Pi

10.5815/ijem.2023.01.03

Cite

Citations (2)

K11. Modified Hybrid Median filter for image denoising

Zeinab A. Mustafa Banazier A. Abrahim Yasser M. Kadah

A critical issue in image restoration is the problem of Gaussian noise removal while keeping the integrity of relevant image information. Clinical magnetic resonance imaging (MRI) data is normally corrupted by Rician noise from the measurement process which reduces the accuracy and reliability of any automatic analysis. The quality of ultrasound (US) imaging is degraded by the presence of signal dependant noise known as speckle. It generally tends to reduce the resolution and contrast, thereby, to degrade the diagnostic accuracy of this modality. For this reasons, denoising methods are often applied to increase the: Signal-to-Noise Ratio (SNR) and improve image quality. This paper proposes a statistical filter, which is a modified version of Hybrid Median filter for noise reduction, which computes the median of the diagonal elements and the mean of the diagonal, horizontal and vertical elements in a moving window and finally the median value of the two values will be the new pixel value. The results show that our proposed method outperforms the classical implementation of the Mean, Median and Hybrid Median filter in terms of denoising quality. Comparison with well established methods, such as Total Variation, Wavelet and Wiener filters show that the proposed filter produces better denoising results, preserving the main structures and details.

Speckle noise

Non-Local Means

Rician fading

Salt-and-pepper noise

Gaussian Noise

10.1109/nrsc.2012.6208586

Cite

Citations (17)

A Computer-Aided Diagnosis System for Classification of Lung Tumors

Journal of Clinical Engineering (2015)

Alla Sh. Abdalla Islam A. Yusuf Sahar Haj Ali A. Mohammed Meinas A. Mahmoud Zeinab A. Mustafa

Lung cancer is the leading cancer killer throughout the world. Despite the boost in technology that has enhanced diagnostic and clinical developments in the medical field, the accuracy in lung tumor evaluation still remains a comprising issue. This article aims toward creating a diagnosis system using artificial neural network to classify the lung tumor either to malignant or benign tumor in computed tomography images. The diagnosing system comprises image processing and artificial neural network procedures. Image processing include procedures such as histogram equalization, image filtering, image segmentation. For the classification system, features were extracted from the segmented images and fed to MLP (multilayer Perceptron) neural network that uses backpropagation algorithm for the learning of the network. Results have rendered the proposed techniques promising with accurate levels of lung cancer detection. The system was able to achieve an accuracy of 95.2% sensitivity, 100% specificity, and an overall classification accuracy of 97.3%. A user-friendly MATLAB graphical user interface program has been constructed to test the proposed algorithm.

Backpropagation

Computer-Aided Diagnosis

Graphical user interface

Multilayer perceptron

10.1097/jce.0000000000000098

Cite

Citations (3)

Brain Tumor Classification Using Principal Component Analysis and Artificial Neural Network

Journal of Clinical Engineering (2019)

Sheroug A. Abdalla Zeinab A. Mustafa Banazier A. Abrahim

The abnormal growth of cells in the brain is known as brain tumor. A brain tumor is a kind of disease that can hit children, adults, and older adults. In this work, a proposed method for brain tumor detection and classification using MATLAB and based on magnetic resonance imaging plays an essential role in the brain-tumor disease diagnostic application that is based on manual and automatic detection. Moreover, various kinds of tumors exist so it is complicated to detect, and thus it is hard to make decisions. Correct segmentation and image enhancement give an accurate classification of brain tumor types. A probabilistic neural network was applied for classification. Two steps were used for making the correct decision: first is feature extraction based on principal component analysis, and second is the classification done using a probabilistic neural network. The known classifications are “normal,” “benign,” and “malignant.”

Brain tumor

Feature (linguistics)

Contextual image classification

10.1097/jce.0000000000000333

Cite

Citations (2)

A Comparative Study of the Effectiveness of Immediate Versus Delayed Photobiomodulation Therapy in Reducing the Severity of Postoperative Inflammatory Complications

Photomedicine and Laser Surgery (2015)

Hala M. Abdel-Alim Hassan Abdel-Dayem Zeinab A. Mustafa Amr Bayoumi Ahmed Jan

Objective: The aim of this study was to compare the immediate versus the delayed application of photobiomodulation (PBM) therapy following odontectomy of horizontally impacted mandibular third molars, and assess which application method is more effective at reducing postoperative complications. Background data: Surgical removal of horizontally impacted mandibular third molars is a common surgical procedure, usually associated with postoperative complications such as pain, swelling, and trismus. Several attempts have been made to minimize these complications. One such method is the use of PBM therapy. Methods: Eighty patients with horizontally impacted mandibular third molars with no inferior alveolar canal approximation were recruited for this study. They were divided into two groups. The immediate group received PBM therapy immediately after surgery and on the 3rd day postoperatively. Subjects in the delayed group received PBM therapy on the 2nd and 4th days postoperatively. All subjects received 2 min of treatment using a 4 W laser beam, during which 171 J were delivered via a 2.8 cm2 spot size. Results: Clinical and statistical results showed a significant reduction in pain, trismus, and swelling in the immediate PBM therapy group compared with the delayed PBM therapy group. Conclusions: Immediate PBM therapy is more effective than delayed PBM therapy in minimizing the complications associated with mandibular third molar removal surgery.

10.1089/pho.2015.3923

Cite

Citations (20)

Modified non-local means filter for effective speckle reduction in ultrasound images

Banazier A. Abrahim Zeinab A. Mustafa Yasser M. Kadah

Ultrasound imaging is a widely used and safe medical diagnostic technique, due to its noninvasive nature, low cost, capability of forming real time imaging, and the continuing improvements in image quality. However; the usefulness of ultrasound imaging is degraded by the presence of signal dependant noise known as speckle. It is well-known that speckle is a multiplicative noise that degrades the visual evaluation in ultrasound imaging. In ultrasound (US) imaging, denoising is intended to improve quantitative image analysis techniques. In this paper, a new version of the Non Local (NL-) means filter adapted for US images is proposed based on Similarity function depend on specific characteristics of the variance speckle noise in ultrasound images. The proposed method has been compared with Median, Wavelet, Mean and variance local statistics, Geometric, Anisotropic diffusion filtering, and Non ' local means filter using quantitative parameters. From the visual results and image quality evaluation metrics obtained over real images we can conclude that the modified(NL-) means filter can be successfully used for ultrasound image denoising, and performs better results than all other methods while still retaining the structural details and retains the edges and textures very well while removing speckle noise.

Speckle noise

10.1109/nrsc.2011.5873623

Cite

Citations (9)