Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Abdullahi Umar Ibrahim has a total of 56 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    The Effect of Variations in Temperature and Contact Time of Zingerone, [6]-Gingerol and Shogaol as Disinfectants on Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa and Candida albicans

    by Fathy A. A. Hasan, Abdullahi Umar Ibrahim, Kaya Suer, Suleyman Asir

    Rising microbial resistance to synthetic disinfectants has intensified the search for chemically synthesized natural alternatives, such as ginger-derived bioactive compounds. Several bioactive compounds, including Zingerone, Gingerols, and Shogaols, have been shown to possess antimicrobial activities. However, the antimicrobial efficacy of these compounds as disinfectants at varying temperatures and contact times is poorly understood. Therefore, understanding the temperature- and time-dependent effects of contact is crucial for optimizing the potential application of these compounds in various antimicrobial strategies. In this study, the antimicrobial activities of three chemicals, 10% [6]-Gingerol, Zingerone, and Shogaols, were evaluated against Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, and Candida albicans. Furthermore, the variations in temperatures (5 °C, 25 °C, and 37 °C) and contact time (1, 5, and 60 min) were assessed based on CFU counts, log10 reductions, percent kill, and decimal reduction time. The findings of this study indicated that 10% Zingerone completely inactivated all tested organisms in 60 min at all temperatures. Although 10% [6]-Gingerol and Shogaol exhibited temperature- and time-dependent effects, they failed to completely inactivate the bacteria and fungi after 60 min. Furthermore, both temperature and contact time were shown to influence the efficacy of the tested disinfectants, providing a significant time- and temperature-dependent reduction in viable cells across all tested organisms.


  • LAG-U-Net: A Lightweight Attention-Guided Network for Efficient Cardiac Ultrasound Segmentation with Cross-Dataset Evaluation

    by Nasim Ahmadzadeh Nobari Azar, Abdullahi Umar Ibrahim, Ammara Talib, Kamil Dimililer, Mohamad Ziad Altabel

    Accurate cardiac structure segmentation from echocardiographic images remains challenging because of speckle noise, low contrast, anatomical variability, and differences in image acquisition. At the same time, many high-performing segmentation networks impose substantial computational requirements. This study presents LAG-U-Net, a lightweight attention-guided encoder-decoder architecture designed to examine the trade-off between segmentation accuracy and computational efficiency in echocardiography. The model uses a compact U-Net-style backbone with attention-guided skip connections and was evaluated for four-class segmentation of the background, left-ventricular cavity, myocardium, and left atrium on the CAMUS dataset. Experiments used the official patient-level training, validation, and test partitions, synchronized training-only augmentation, class-aware loss optimization, multiple CNN and transformer-based baselines, and three independent random seeds. Across the three runs, LAG-U-Net achieved a mean foreground Dice Similarity Coefficient (DSC) of 0.8717 ± 0.0013 and an Intersection-over-Union (IoU) of 0.7837 ± 0.0017. Its DSC was slightly lower than that of conventional U-Net (0.8769 ± 0.0024) but slightly higher than that of the parameter-matched LiteUNet (0.8687 ± 0.0025). LAG-U-Net required 1.95 million trainable parameters compared with 7.79 million for U-Net and achieved an NVIDIA L4 inference latency of 3.38 ms per image compared with 4.92 ms for U-Net, demonstrating a favorable accuracy-efficiency trade-off. External evaluation was additionally performed on the left-ventricular class of EchoNet-Dynamic without fine-tuning. DSC decreased from 0.9201 ± 0.0008 for the CAMUS left-ventricular class to 0.1578 ± 0.0087 on EchoNet-Dynamic, indicating substantial cross-dataset domain shift and limited zero-shot transfer. Overall, LAG-U-Net provides a compact segmentation architecture with competitive within-dataset performance and reduced computational requirements, while the external results emphasize the need for domain adaptation and broader multi-center validation before clinical deployment.

  • Open Access

    Nano-modified biosensors for detection of pathogenic diseases: The prospect of smart, multiplex and point-of-care testing

    by Abdullahi Umar Ibrahim, Pwadubashiyi Coston Pwavodi, Mehmet Ozsoz, Basil Birth Duwa, Irkam Irkham, Yeni Wahyuni Hartati

    Introduction and Background: The world has witnessed several outbreaks, emergence and re-emergence of infectious diseases throughout the 21st century as a result of climate change, urbanization and migration. Several infectious diseases caused by pathogens such as SARS-CoV-2, Ebola, Zika, Dengue, Marburg viruses, Mycobacterium tuberculosis, etc. have caused a devastating impact on lives and livelihoods around the world. To counter these diseases, medical experts rely on conventional techniques, which include microscopy and serological testing. However, these conventional methods are hindered by several trade-offs, including high cost, longer processing times, low sensitivity, and a likelihood of false positive results. Biomedical sensors have gained momentum in clinical diagnostics due to their low cost, portability, and sensitivity, among other advantages. To improve their performance, scientists have incorporated nanomaterials. Other techniques used to enhance the performance of nanobiosensors include multiplex testing, point-of-care testing (POCT), and smart sensing. Methodology: Thus, in this review, we present a comprehensive overview of the state-of-the-art nanobiosensors for detecting infectious diseases. The review covers key topics which centred around the application of nanotechnology in biosensing, multiplex testing, POCT and smart nano-enhanced biosensors. Findings: The findings of this review highlighted the advantages of biosensors over conventional approaches, with a limit of detection ranging from nanomolar to attomolar concentrations and a time response ranging from 1 to 3 hours. Conclusion: Despite the prospect of nanobiosensors, several limitations such as complexity, very extensive, longer processing time, etc. Moreover, the integration of smart technologies in nanobiosensors can offer several benefits, which include high accuracy, faster detection and prediction.


  • MODELING AND STABILITY ANALYSIS OF NEUROLOGICAL DISORDER UNDER ALZHEIMER’S DISEASE WITH NOVEL FRACTIONAL TECHNIQUE

    by Ammara Talib, Muhammad Farman, Abdullahi Umar Ibrahim, Khadija Jamil, Kottakkaran Sooppy Nisar, Aceng Sambas

    In this study, we developed a novel fractional-order mathematical model to understand the dynamics of Alzheimer’s disease progression based on different factors by using the Caputo fractional derivative. The framework is applied to simulate key pathological markers, such as amyloid-beta accumulation, tau protein hyperphosphorylation, and neuronal degeneration. The uniqueness and exitance of the solution in this study are mathematically supported by the Banach contraction principle. Stability and controllability analyses are performed, including the implementation of a fractional-order PID controller to manage chaos and ensure the system’s equilibrium. Numerical simulations across fractional orders reveal that lower fractional values significantly delay the buildup of [ ], dampen neurodegeneration, and slow cognitive decline. The simulations reveal the model’s capability to capture the complex dynamics of Alzheimer’s disease progression, offering insights into its nonlinear characteristics and potential therapeutic interventions. The findings support personalized treatment planning through data-driven computational approaches, particularly in light of newly approved disease-modifying therapies.


  • Open Access

    I-Brainer: Artificial intelligence/Internet of Things (AI/IoT)-Powered Detection of Brain Cancer

    by Abdullahi Umar Ibrahim, Ikedichukwu Onyemaucheya Nwaneri, Mercel Vubangsi, Fadi Al-Turjman

    Background/Objective: Brain tumor is characterized by its aggressive nature and low survival rate and therefore, it is regarded as one of the deadliest diseases. Thus, misdiagnosis or miss-classification of brain tumors can lead to miss-treatment or incorrect treatment and reduce survival chances. Therefore, there is a need to develop a technique that can identify and detect brain tumors at early stages. Methods: Here, we proposed a framework titled I-Brainer which is an Artificial Intelligence/Internet of Things (AI/IoT)-powered classification of MRI into 4 classes. We employed a Br35H+SARTAJ brain MRI dataset which contains 7023 total images including no tumor, pituitary, meningioma, and glioma. To accurately classify MRI into 4 classes, we developed the LeNet model from scratch, and implemented 2 pre-trained models which include EfficientNet and ResNet-50 as well as feature extraction of these models coupled with 2 Machine Learning (ML) classifiers namely; k- Nearest Neighbours (KNN) and Support Vector Machine (SVM). Results: Evaluation and comparison of the performance of the 3 models have shown that ResNet-50 achieved the best result in terms of AUC (99%) and ResNet-50-KNN ranked higher in terms of accuracy (94%) on the testing set. Conclusion: This framework can be harnessed by patients residing in remote areas and as a confirmatory approach for medical experts.


  • I-BrainNet: Deep Learning and Internet of Things (DL/IoT)–Based Framework for the Classification of Brain Tumor

    by Abdullahi Umar Ibrahim, Glodie Mpia Engo, Ibrahim Ame, Chidi Wilson Nwekwo, Fadi Al-Turjman

    Brain tumor is categorized as one of the most fatal form of cancer due to its location and difficulty in terms of diagnostics. Medical expert relies on two key approaches which include biopsy and MRI. However, these techniques have several setbacks which include the need of medical experts, inaccuracy, miss-diagnosis as a result of anxiety or workload which may lead to patient morbidity and mortality. This opens a gap for the need of precise diagnosis and staging to guide appropriate clinical decisions. In this study, we proposed the application of deep learning (DL)–based techniques for the classification of MRI vs non-MRI and tumor vs no tumor. In order to accurately discriminate between classes, we acquired brain tumor multimodal image (CT and MRI) datasets, which comprises of 9616 MRI and CT scans in which 8000 are selected for discrimination between MRI and non-MRI and 4000 for the discrimination between tumor and no tumor cases. The acquired images undergo image pre-processing, data split, data augmentation and model training. The images are trained using 4 DL networks which include MobileNetV2, ResNet, Ineptionv3 and VGG16. Performance evaluation of the DL architectures and comparative analysis has shown that pre-trained MobileNetV2 achieved the best result across all metrics with 99.94% accuracy for the discrimination between MRI and non-MRI and 99.00% for the discrimination between tumor and no tumor. Moreover, I-BrainNet which is a DL/IoT-based framework is developed for the real-time classification of brain tumor.


  • Open Access

    Computer-Aided/Internet of Medical Things (CAD/IoMT)-Powered Detection of Brain Tumor

    by Misbahu Koramar Boko, Abdullahi Umar Ibrahim, Fadi Al-Turjman, Pwadubashiyi Coston Pwavodi

    Background/Objectives: Brain tumor continue to cause concern globally due to increasing number of cases and mortality. As the first line of action, accurate and early screening of brain tumor is critical for proper treatment and extending the life span of patients. Currently, medical expert relies heavily on the use of magnetic resonance imaging (MRI) for the detection of brain tumor. However, one of the limitation of MRI revolves around manual interpretation, which is time consuming and can be prone to errors especially when dealing with large number of cases. Thus, in order to address this issue, we proposed the development of CAD/IoMT-powered platform that enables real-time and fast detection of brain tumor. Methods: We proposed a framework known as I-BRAIN-DETECT, which is a CAD-based system, integrated with IoMT for fast and real-time detection of brain tumors and no tumor from MRIs. The overall methodology revolves around the use of 2 publicly accessible datasets, image pre-processing, feature extraction and classification using untrained customized CNN and 5 pre-trained CNNs which include ResNet-18, ResNet-50, DenseNet121, MobileNetV2 and EfficientNetB0. Results: Performance evaluation and comparative between implemented models has shown that pre-trained ResNet18 achieve the best result with 98.83% accuracy, 98.33 % recall, 99.33% precision, 99.33% specificity, 98.83% F1-score and 99.92 AUC for binary classification, while MobileNetV2 achieved the best result with 92.93% accuracy, 92.93 % recall, 93.37% precision, 97.67% specificity, 92.79% F1-score and 100 AUC for multiclass-classification. Conclusions: The developed platform can now be access by both patients and medical experts for real-time screening of brain tumor.

  • Open Access

    Advancements in synthesis of Molecularly Imprinted Polymer (MIPs) for highly selective alcohol sensors

    by Wulan Khaerani, Abdullahi Umar Ibrahim, Uji Pratomo, Souvia Rahimah, Irkham Irkham, Yeni Wahyuni Hartati


  • Open Access

    Review of: "Application of Ensemble Learning in CXR Classification for Improving COVID-19 Diagnosis"

    by Abdullahi Umar Ibrahim

    Title: Application of Ensemble Learning in CXR Classification for Enhancing COVID-19 Diagnosis Comments It has been almost 2 years since the end of the COVID-19 pandemic. Several studies have developed thousands of CAD-based models. Therefore, this...