Afeez Adekunle Soladoye

Mr. Afeez Adekunle Soladoye

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 August 2025 in International Journal of Medical Informatics
Verified
Journal Article
Open access
ASAfeez Adekunle Soladoye
Afeez Adekunle Soladoye
NANicholas Aderinto
Nicholas Aderinto
DODamilola Osho
Damilola Osho
DODavid Bamidele Olawade
David Bamidele Olawade
Alzheimer's disease (AD) represents a significant global health challenge requiring early and accurate prediction for effective intervention. While machine learning models demonstrate promising capabilities in AD prediction, their black-box nature limits clinical adoption due to a lack of interpretability and transparency. This study aims to develop and evaluate explainable artificial intelligence (XAI) frameworks for AD prediction using comprehensive multimodal patient data, with a focus on enhancing model interpretability through SHAP and LIME techniques. A comprehensive dataset of 2,149 patients aged 60-90 years was obtained from Kaggle, encompassing demographic, medical history, lifestyle, clinical measurements, cognitive assessments, and symptom data. Rigorous preprocessing included MinMax normalisation, Synthetic Minority Over-sampling Technique (SMOTE) for class imbalance, and Backward Elimination Feature Selection reduced 32 features to 26 optimal predictors. Six machine learning models were evaluated: K-Nearest Neighbours (KNN), Support Vector Machine (SVM), Logistic Regression (LR), XGBoost, Stacked Ensemble, and Random Forest (RF). RF's optimal hyperparameters were obtained using Ant colony Optimization Model interpretability was enhanced using SHAP and LIME frameworks for both global and local explanations. The optimised Random Forest with backward elimination feature selection and ant colony optimisation achieved superior performance with 95 % accuracy, 95 % precision, 94 % recall, 94 % F1-score, and 98 % AUC. SHAP analysis identified functional assessment, activities of daily living (ADL), memory complaints, and Mini-Mental State Examination (MMSE) as the most influential predictors. LIME provided complementary local explanations, validating the clinical relevance of identified features. The integration of explainable AI techniques with machine learning models provides clinically meaningful insights for AD prediction, enhancing transparency and fostering trust in AI-driven diagnostic tools whilst maintaining high predictive accuracy. Future work should focus on external validation, clinical workflow integration, and addressing computational requirements for real-world deployment.

5 September 2025 in Current Research in Translational Medicine
Verified
Journal Article
Open access
AEAdebimpe O. Esan
Adebimpe O. Esan
DODavid Bamidele Olawade
David Bamidele Olawade
ASAfeez Adekunle Soladoye
Afeez Adekunle Soladoye
BOBolaji A. Omodunbi
Bolaji A. Omodunbi
IAIbrahim A. Adeyanju
Ibrahim A. Adeyanju
NANicholas Aderinto
Nicholas Aderinto
Parkinson's Disease (PD) is a chronic, progressive neurological disorder with significant clinical and economic impacts globally. Early and accurate prediction remains challenging with traditional diagnostic methods due to subjectivity, delayed diagnosis, and variability. Machine Learning (ML) approaches offer potential solutions, yet their clinical adoption is hindered by limited interpretability. This study aimed to develop an interpretable ML model for early and accurate PD prediction using comprehensive multimodal datasets and Explainable Artificial Intelligence (XAI) techniques. The study applied five ML algorithms: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Logistic Regression (LR), Random Forest (RF), XGBoost, and a stacked ensemble method to a publicly available dataset (n = 2105) from Kaggle. Data encompassed demographic, medical history, lifestyle, clinical symptoms, cognitive, and functional assessments with specific inclusion/exclusion criteria applied. Preprocessing involved normalization, Synthetic Minority Oversampling Technique (SMOTE), and Sequential Backward Elimination (SBE) for feature selection. Model performance was evaluated via accuracy, precision, recall, F1-score, and Area Under Curve (AUC). The best-performing model (RF with feature selection) was interpreted using SHAP and LIME methods. Random Forest combined with Backward Elimination Feature Selection achieved the highest predictive accuracy (93 %), precision (93 %), recall (93 %), F1-score (93 %), and AUC (0.97). SHAP and LIME analyses indicated UPDRS scores, cognitive impairment, functional assessment, and motor symptoms as primary predictors, enhancing clinical interpretability. The study demonstrated the effectiveness of an interpretable RF model for accurate PD prediction. Integration of ML and XAI significantly improves clinical decision-making, diagnosis timing, and personalized patient care.

27 January 2025 in Clinical Neurology and Neurosurgery
Verified
Journal Article
Open access
ASAfeez Adekunle Soladoye
Afeez Adekunle Soladoye
DODavid Bamidele Olawade
David Bamidele Olawade
IAIbrahim A. Adeyanju
Ibrahim A. Adeyanju
OAOnoja M. Akpa
Onoja M. Akpa
NANicholas Aderinto
Nicholas Aderinto
MOMayowa O. Owolabi
Mayowa O. Owolabi
Stroke remains a leading cause of death and disability worldwide, with African populations bearing a disproportionately high burden due to limited healthcare infrastructure. Early prediction and intervention are critical to reducing stroke outcomes. This study developed and evaluated a stroke prediction system using Gated Recurrent Units (GRU), a variant of Recurrent Neural Networks (RNN), leveraging the Afrocentric Stroke Investigative Research and Education Network (SIREN) dataset. The study utilized secondary data from the SIREN dataset, comprising 4236 records with 29 phenotypes. Feature selection reduced these to 15 optimal phenotypes based on their significance to stroke occurrence. The GRU model, designed with 128 input neurons and four hidden layers (64, 32, 16, and 8 neurons), was trained and evaluated using 150 epochs, a batch size of 8, and metrics such as accuracy, AUC, and prediction time. Comparisons were made with traditional machine learning algorithms (Logistic Regression, SVM, KNN) and Long Short-Term Memory (LSTM) networks. The GRU-based system achieved a performance accuracy of 77.48 %, an AUC of 0.84, and a prediction time of 0.43 seconds, outperforming all other models. Logistic Regression achieved 73.58 %, while LSTM reached 74.88 % but with a longer prediction time of 2.23 seconds. Feature selection significantly improved the model's performance compared to using all 29 phenotypes. The GRU-based system demonstrated superior performance in stroke prediction, offering an efficient and scalable tool for healthcare. Future research should focus on integrating unstructured data, validating the model on diverse populations, and exploring hybrid architectures to enhance predictive accuracy.

9 June 2025 in Diagnostics
Verified
Journal Article
Open access
PDF file
BOBolaji A. Omodunbi
Bolaji A. Omodunbi
DODavid Bamidele Olawade
David Bamidele Olawade
Omosigho Florence AweOmosigho Florence Awe
Omosigho Florence Awe
ASAfeez Adekunle Soladoye
Afeez Adekunle Soladoye
NANicholas Aderinto
Nicholas Aderinto
Stergios BoussiosStergios Boussios
Stergios Boussios
Background: Parkinson’s disease (PD) is a progressive neurodegenerative condition that impairs motor and non-motor functions. Early and accurate diagnosis is critical for effective management and care. Leveraging machine learning (ML) techniques, this study aimed to develop a robust prediction system for PD using a stacked ensemble learning approach, addressing challenges such as imbalanced datasets and feature optimization. Methods: An open-access PD dataset comprising 22 vocal attributes and 195 instances from 31 subjects was utilized. To prevent data leakage, subjects were divided into training (22 subjects) and testing (9 subjects) groups, ensuring no subject appeared in both sets. Preprocessing included data cleaning and normalization via min–max scaling. The synthetic minority oversampling technique (SMOTE) was applied exclusively to the training set to address class imbalance. Feature selection techniques—forward search, gain ratio, and Kruskal–Wallis test—were employed using subject-wise cross-validation to identify significant attributes. The developed system combined support vector machine (SVM), random forest (RF), K-nearest neighbor (KNN), and decision tree (DT) as base classifiers, with logistic regression (LR) as the meta-classifier in a stacked ensemble learning framework. Performance was evaluated using both recording-wise and subject-wise metrics to ensure clinical relevance. Results: The stacked ensemble learning model achieved realistic performance with a recording-wise accuracy of 84.7% and subject-wise accuracy of 77.8% on completely unseen subjects, outperforming individual classifiers including KNN (81.4%), RF (79.7%), and SVM (76.3%). Cross-validation within the training set showed 89.2% accuracy, with the performance difference highlighting the importance of proper validation methodology. Feature selection results showed that using the top 10 features ranked by gain ratio provided optimal balance between performance and clinical interpretability. The system’s methodological robustness was validated through rigorous subject-wise evaluation, demonstrating the critical impact of validation methodology on reported performance. Conclusions: By implementing subject-wise validation and preventing data leakage, this study demonstrates that proper validation yields substantially different (and more realistic) results compared to flawed recording-wise approaches. The findings underscore the critical importance of validation methodology in healthcare ML applications and provide a template for methodologically sound PD classification research. Future research should focus on validating the model with larger, multi-center datasets and implementing standardized validation protocols to enhance clinical applicability.

29 July 2025 in Discover Data
Verified
Journal Article
Open access
PDF file
ASAfeez Adekunle Soladoye
Afeez Adekunle Soladoye
KOKazeem Michael Olagunju
Kazeem Michael Olagunju
Sunday Adeola AJAGBESunday Adeola AJAGBE
Sunday Adeola AJAGBE
IAIbrahim A. Adeyanju
Ibrahim A. Adeyanju
POPrecious I. Ogie
Precious I. Ogie
PMPragasen Mudali
Pragasen Mudali
The application of Artificial Intelligence (AI) to diverse field has been widely accepted ranging from transportation, education, logistics, entertainment and health. Specifically, in recent time, the application of Machine Learning (ML) a subset of AI has equally got wide acceptance and relevance in various aspect of medicine ranging from diagnosis and prediction of diseases, development of drugs and treatment plan among others. This has made medical procedure to be faster, more accurate and easier than using the traditional approach. Stroke is reported to be one of the major causes of death and this can be reduced by studying the risk factors causing it and predicting its occurrence so as to educate people about it. This study developed a stroke prediction system with a modified Gated Recurrent Unit (GRU), a structured stroke dataset was gotten from Kaggle, which went through different preprocessing techniques like label Encoder, Normalization with MinMax, dropping of irrelevant values. Furthermore, different data balancing techniques were employed to improve the accurate performance of the model. The preprocessed dataset was used by GRU for prediction. The system gave average accuracy, Area Under Curve (AUC) and prediction time of 80.42%, 0.8940 and 0.678 s respectively. The developed system outperformed other ML algorithms like LSTM, GRU-LSTM, Support Vector Machine (SVM) and Logistic Regression. This study showed that GRU a variant of Recurrent Neural Network, could give better predictive performance on structured data rather than only with streaming data; therefore, showcasing the improved performance of GRU over other variants of RNN.

1 May 2025 in ABUAD Journal of Engineering Research and Development (AJERD)
Verified
Journal Article
Open access
PDF file
OAOpeyemi Oreoluwa Asaolu
Opeyemi Oreoluwa Asaolu
OAOluwasanmi Segun Adanigbo
Oluwasanmi Segun Adanigbo
ASAfeez Adekunle Soladoye
Afeez Adekunle Soladoye
NONnamdi Stephen Okomba
Nnamdi Stephen Okomba
The increasing occurrence of Distributed Denial of Service (DDoS) attacks has caused significant disruptions in global network services, overwhelming targets by flooding them with requests from various sources. This ease of execution and gaining entry to distributed systems for rent has led to increasing financial losses. This paper addresses the growing challenge of IoT devices-targeted Distributed Denial of Service (DDoS) attacks within 4G networks. In this study, a PCA-MLP (Principal Component Analysis-Multi-Layer Perceptron) intrusion detection model combined with a packet-filtering firewall for enhanced prevention is presented. The firewall, utilizing IPtables, selectively permits traffic from trusted sources, successfully blocking nearly 70% of DDoS threats. The PCA-MLP model proposed in this study demonstrated high performance, accurately identifying different types of DDoS attacks with an overall accuracy of 95.35%.