Scholarly Publication

Permanent URI for this collection

https://kwasuspace.kwasu.edu.ng/handle/123456789/48

Browse

Recent Submissions

Now showing 1 - 5 of 284
  • Thumbnail Image
    Item
    Anomaly Detection in Social Media Conversation Using Natural Language Processing with LSTM and Naïve Bayes Models
    (Academic City University College, Accra, Ghana, 2024) Babatunde, A. N., Shuaib, B. M., Kadri, A. F., Isiaka, O. S., Abdulrahman, T. A., Ismail, S. I. & Oke, A. A. (2024)
  • Thumbnail Image
    Item
    Joint segmentation and classification of cervical cells using U-Net and CNN ensemble for early detection of cervical cancer. Cancer Cell International
    (Web of Science, 2026) 1. Babatunde, A. N., Aroba, O. J., Adeniyi, A. E., Isiaka, S. O., Abdulrahman, T. A. Kadri, A. F. Mohammed, S. B. & Rudolph, M. (2026)
    Cervical cancer remains a leading cause of mortality among women, particularly in resource-limited settings where cytological screening is hindered by manual analysis challenges. This study presents a joint segmentation and classification framework integrating U-Net and a CNN ensemble to automate and enhance the accuracy of cervical cell analysis. The U-Net architecture is employed to delineate nuclei and cytoplasmic boundaries from Pap smear images, leveraging skip connections to preserve spatial resolution in complex cellular environments. Following segmentation, an ensemble classifier combining ResNet-50, VGG-16, and InceptionV3 is utilised to differentiate cervical cell types with high precision. The proposed method was evaluated on benchmark datasets (APACS-23, Cx22, SIPaKMeD) using stratified five-fold cross-validation and external validation cohorts. Experimental results achieved a mean Dice coefficient of 0.934 for segmentation and a classification accuracy of 94.6%, outperforming single-model baselines by 5–8% in F1-score and AUC metrics. Grad-CAM visualisations demonstrated that the model’s focus corresponded to diagnostically relevant regions, reinforcing clinical interpretability. Ablation studies confirmed that the joint learning architecture substantially enhanced feature localisation and classification robustness. This research contributes a deployable, interpretable AI-based pipeline suitable for real-time cervical cancer screening, offering potential deployment in both centralised laboratories and low-resource healthcare environments. Future directions include expanding dataset diversity, integrating transformer-based models, and developing lightweight implementations for edge devices. Keywords Cervical Cancer Screening, Joint Segmentation and Classification, U-Net, CNN Ensemble, Medical Image Analysis, Explainable AI
  • Thumbnail Image
    Item
    A Lightweight Hybrid MobileNetV3-Random Forest Model for Diabetic Foot Ulcer Detection for Resource-Constrained Nigerian Clinics
    (Department of Computer Science, Kwara State University, Malete & Al-Hikmah University, Ilorin, Nigeria, 2026) Lawal, A. O., Babatunde, A. N., Babatunde, R. S., Lawal, I. A., Kadri, A. F., Muhammed, S. B. & Ekundayo, O. (2026)
    : In Nigeria, Diabetic Foot Ulcer (DFU) is a severe consequence of diabetes that causes high rates of lower-leg amputations and death. Many primary care clinics throughout the nation depend on subjective, non-quantitative clinical judgments, which frequently lead in delayed diagnosis. The lightweight hybrid MobileNetV3-Random Forest model created in this study for automated DFU detection from smartphone-captured pictures, optimized for inexpensive devices found in resource-limited Nigerian clinics, confronts this problem. From several Nigerian medical sources, 1,721 anonymized DFU photographs including a variety of skin tones and ulcer grades were compiled. To increase robustness against lighting and orientation variances, images were normalized, resized to 224x224 pixels, and augmented. Reducing the image dimensionality from 28,224 to 96 principal components while retaining 95% of the variance, PCA was then utilized. The feature extractor, MobileNetV3 (pre-trained through transfer learning), generated a high dimensional feature vector. For the ultimate DFU detection job, these condensed elements were input into a Random Forest classifier. With five-fold cross-validation employed to guarantee robustness, the dataset was split into 70/20/10 for training, testing and cross-validation respectively. Achieving 92.8% accuracy, 0.93 precision, 0.93 recall, 0.93 F1-score, and 0.9959 ROC-AUC, the hybrid model outperformed several heavier benchmarks in efficiency. Ablation studies validated that PCA decreased memory use by 31% on low-end devices and reduced inference time by about 47% (from 352 ms to 187 ms per image), with very little influence on predictive performance. This study offers a scalable, computationally economical, and understandable artificial intelligence framework supporting hybrid intelligence ideas that advances fair early diagnosis of diabetic foot ulcers in resource-poor medical environments. KEYWORD: Diabetic Foot Ulcer (DFU), Artificial Intelligence, Principal Component Analysis, Convolutional Neural Network, MobileNetV3, Random Forest Algorithm.
  • Thumbnail Image
    Item
    A Real-Time Gender and Age Prediction System Based on Facial Images Using Convolutional Neural Networks
    (Faculty of Engineering and Computing, University of Science & Technology, Aden, Yemen, 2025) Mohammed, S. B., Olajide, A. T., Kadri, A. F., Ilori, O., Tajudeen, K. O., Sulaiman, H. O., Ekundayo, O. & Babatunde, A. N. (2025)
    Real-time gender and age prediction based on facial images have become essential in applications such as surveillance, demographic analytics, and human-computer interaction. However, existing systems often struggle with challenges like inconsistent lighting, occlusions, pose variations, and dataset imbalances, which reduce accuracy and hinder deployment on resource-constrained devices. This research introduces a real-time gender and age prediction system that leverages Convolutional Neural Networks (CNNs), utilizing fine-tuned ResNet-34 for gender classification and a modified VGG-16 for age estimation. A robust preprocessing pipeline, integrating image normalization, histogram equalization, data augmentation, and synthetic data generation, enhances model generalization across diverse datasets. Efficient face detection techniques, combined with lightweight model architectures, enable the system to achieve inference speeds below 500 milliseconds per frame without GPU acceleration. Experimental evaluations across IMDB-WIKI, UTKFace, MORPH II, and Adience datasets revealed gender classification accuracies exceeding 94% and age prediction one-off accuracies above 90%. The proposed system demonstrates a scalable and efficient solution for real-time facial attribute recognition in dynamic and resource-limited environments. Keywords— Gender Classification, Age Estimation, Convolutional Neural Networks, Real-Time Facial Recognition, Data Augmentation, Cross-Dataset Generalization
  • Thumbnail Image
    Item
    Hybrid CNN–Decision Tree Framework for Medicinal Leaf Recognition
    (Faculty of Science and Technology, Federal University Wukari, Taraba State, Nigeria, 2026)
    Accurate and automated identification of medicinal plant species is crucial for various applications, including pharmaceutical research, agricultural practices, and the preservation of ethnobotanical knowledge. Traditional manual identification methods are often time-consuming, prone to human error, and require specialized taxonomic expertise, limiting their scalability and accessibility. This study addresses these challenges by developing and comparatively evaluating two machine learning approaches Convolutional Neural Networks (CNNs) and Decision Trees (DT) for the recognition of medicinal plant leaves. Our proposed system leverages deep learning techniques to analyze visual features of leaf images, enabling efficient and precise classification. The research involved a comprehensive process of dataset collection, meticulous preprocessing, model training, and rigorous performance evaluation using standard metrics such as accuracy, precision, recall, and F1-score. We incorporated both automated feature extraction by CNNs and handcrafted features, including Gray Level Co-occurrence Matrix (GLCM) for Decision Trees, to provide a robust comparative analysis. The CNN model demonstrated superior performance, achieving an accuracy of 97.2% compared to the Decision Tree's 85.6%, alongside higher precision, recall, and F1-scores. This highlights the CNN's robustness and its capability to learn intricate hierarchical features from complex visual data. The findings suggest that CNNs are highly suitable for practical deployment in real-world scenarios, contributing significantly to improved accessibility, preservation, and application of traditional medicinal knowledge through advanced automated identification systems. Keywords: Medicinal Plant Identification, Leaf image classification, Convolutional Neural Networks (CNN), Decision Tree (DT), Feature Extraction, Image processing