Sumaiya Nezam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Sumaiya Nezam has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Chen-Chih Chung
Department of Neurology, Shuang Ho Hospital, Taipei Medical University, New Taipei City, Taiwan
2 shared publications Donaldson F. Conserve
Affiliation Department of Community Health and Prevention, Milken Institute School of Public Health, George Washington University, Washington, District of Columbia, United States of America
2 shared publications- Oluwaseun Adebayo Bamodu
Department of Prevention and Community Health, Milken Institute School of Public Health, The George Washington University, Washington, DC 20052, USA
2 shared publications Abdulrazzaq Lawal
Affiliation Department of Surgery, College of Medicine, University of Lagos, Lagos, Nigeria
1 shared publicationAdedotun Ogunbajo
Affiliation Us Helping US, People Into Living, Inc., Washington, DC, United States of America
1 shared publication- Adewumi Alabi
University of Lagos
1 shared publication
Latest Publications
- Open Access
- 11 September 2026
- Published by Public Library of Science (PLoS) in PLOS Digital Health
- Vol. 5 (9) , e0001706
Explainable machine learning for breast cancer prediction in resource-constrained settings: A multi-algorithmic framework integrating shap-based transparency with clinical decision support
Breast cancer remains the most commonly diagnosed malignancy among women globally, with disproportionately higher mortality rates in low- and middle-income countries (LMICs) where diagnostic delays and limited specialist pathology capacity are widespread. While machine learning (ML) approaches achieve strong predictive performance for cancer classification, algorithmic opacity and absence of interpretability frameworks tailored to resource-constrained environments have impeded clinical adoption. This study bridges the translational gap between predictive accuracy and clinical utility by developing an explainable artificial intelligence (XAI) framework specifically designed for breast cancer diagnosis in underserved healthcare settings. Using the Wisconsin Breast Cancer Diagnostic Dataset (569 fine-needle aspirate cytological specimens with 30 nuclear morphometric features), we systematically benchmarked eight supervised classification algorithms: Logistic Regression, Random Forest, XGBoost, LightGBM, Support Vector Machine (SVM), Gradient Boosting, Decision Tree, and K-Nearest Neighbors, using stratified 10-fold cross-validation and an independent hold-out test set (80:20 split). Performance was evaluated across discriminative and probabilistic metrics, including AUC-ROC, F1-score, Matthews Correlation Coefficient (MCC), and Brier score, and interpretability was operationalized through SHapley Additive exPlanations (SHAP) analysis with global feature importance, cross-model consensus ranking, and individual-level dependence characterization. All ensemble and regularized models achieved test-set AUCs above 0.98, with XGBoost and SVM attaining the highest AUC of 0.996, and Logistic Regression the highest accuracy (98.25%) and MCC (0.962). SHAP analysis consistently identified worst perimeter, worst concave points, and worst area as the dominant predictors, with strong concordance across gradient-boosted models (pairwise Spearman rho: XGBoost–LightGBM 0.86, XGBoost–Random Forest 0.82, Random Forest–LightGBM 0.67). Logistic Regression also demonstrated superior probability calibration, a critical requirement for clinical risk stratification. Collectively, these findings deliver a reproducible, transparent framework whose SHAP-derived signatures align with established cytopathological principles, supporting responsible integration of interpretable ML into resource-limited diagnostic workflows and providing a template for equitable AI deployment in global oncology.
- 11 September 2026
- Published by Public Library of Science (PLoS) in PLOS Digital Health
- Vol. 5 (9) , e0001706
Open AccessExplainable machine learning for breast cancer prediction in resource-constrained settings: A multi-algorithmic framework integrating shap-based transparency with clinical decision support
Breast cancer remains the most commonly diagnosed malignancy among women globally, with disproportionately higher mortality rates in low- and middle-income countries (LMICs) where diagnostic delays and limited specialist pathology capacity are widespread. While machine learning (ML) approaches achieve strong predictive performance for cancer classification, algorithmic opacity and absence of interpretability frameworks tailored to resource-constrained environments have impeded clinical adoption. This study bridges the translational gap between predictive accuracy and clinical utility by developing an explainable artificial intelligence (XAI) framework specifically designed for breast cancer diagnosis in underserved healthcare settings. Using the Wisconsin Breast Cancer Diagnostic Dataset (569 fine-needle aspirate cytological specimens with 30 nuclear morphometric features), we systematically benchmarked eight supervised classification algorithms: Logistic Regression, Random Forest, XGBoost, LightGBM, Support Vector Machine (SVM), Gradient Boosting, Decision Tree, and K-Nearest Neighbors, using stratified 10-fold cross-validation and an independent hold-out test set (80:20 split). Performance was evaluated across discriminative and probabilistic metrics, including AUC-ROC, F1-score, Matthews Correlation Coefficient (MCC), and Brier score, and interpretability was operationalized through SHapley Additive exPlanations (SHAP) analysis with global feature importance, cross-model consensus ranking, and individual-level dependence characterization. All ensemble and regularized models achieved test-set AUCs above 0.98, with XGBoost and SVM attaining the highest AUC of 0.996, and Logistic Regression the highest accuracy (98.25%) and MCC (0.962). SHAP analysis consistently identified worst perimeter, worst concave points, and worst area as the dominant predictors, with strong concordance across gradient-boosted models (pairwise Spearman rho: XGBoost–LightGBM 0.86, XGBoost–Random Forest 0.82, Random Forest–LightGBM 0.67). Logistic Regression also demonstrated superior probability calibration, a critical requirement for clinical risk stratification. Collectively, these findings deliver a reproducible, transparent framework whose SHAP-derived signatures align with established cytopathological principles, supporting responsible integration of interpretable ML into resource-limited diagnostic workflows and providing a template for equitable AI deployment in global oncology.
- Open Access
- 31 March 2026
- Published by Public Library of Science (PLoS) in PLOS Global Public Health
- Vol. 6 (3) , e0005969
Depression and cancer outcomes in resource-limited settings: A cross-sectional analysis of treatment uptake and survival among metastatic breast cancer patients in Lagos, Nigeria
The intersection of depression and cancer outcomes represents a critical yet understudied domain in global oncology, particularly in low- and middle-income countries (LMICs) where both conditions impose substantial disease burdens. While evidence from high-income countries demonstrates associations between depression and adverse cancer outcomes, the generalizability of these findings to resource-limited settings remains uncertain. Theoretical frameworks including Beck’s cognitive model and the biopsychosocial model suggest multiple pathways through which depression may influence cancer outcomes; however, these pathways may operate differently in resource-constrained healthcare environments. This study examines the relationship between clinically significant depression (CSD) and treatment uptake and survival among metastatic breast cancer patients in Nigeria, providing novel insights into psycho-oncology within the sub-Saharan African context. We conducted a cross-sectional analysis of 313 metastatic breast cancer patients presenting to NSIA-LUTH Cancer Centre, Lagos, Nigeria (September 2020–February 2022). Depression was assessed using the Beck Depression Inventory-II (BDI-II), with CSD defined as BDI-II ≥ 20, consistent with oncology-validated cutoffs and African psychometric validation studies. Associations between CSD and treatment receipt were evaluated using multivariable logistic regression. Survival patterns were examined using Kaplan-Meier analysis and Cox proportional hazards models, adjusting for clinicopathological confounders. The cross-sectional design precludes causal inference regarding temporal relationships between depression and outcomes. Among 274 patients with complete data (mean age 54.0 years, 52.1% triple-negative), depression was prevalent (52.6% any depression; 17.9% CSD). Treatment uptake was 64.6% overall. Contrary to findings from high-income settings, CSD showed no significant association with treatment receipt (adjusted OR 0.95, 95% CI 0.36–2.52, p = 0.92) or mortality (adjusted HR 1.12, 95% CI 0.48–2.61, p = 0.80). Clinical factors dominated outcome prediction: lung metastasis (OR 3.63), ulceration (OR 20.39), and lymph node status significantly predicted treatment receipt, while multimodal therapy was strongly protective against mortality (HR 0.25, 95% CI 0.10–0.65). Findings should be interpreted in the context of 12.5% missing depression data and potential selection and misclassification biases. In this Nigerian cohort, CSD was common but not associated with treatment uptake or survival. These findings suggest that in resource-constrained settings where clinical triage predominates, tumor biology and treatment access may supersede psychosocial factors as outcome determinants. The substantial depression prevalence nonetheless warrants integrated mental health services within cancer care, with implications for comprehensive oncology models in LMICs.
- 31 March 2026
- Published by Public Library of Science (PLoS) in PLOS Global Public Health
- Vol. 6 (3) , e0005969
Open AccessDepression and cancer outcomes in resource-limited settings: A cross-sectional analysis of treatment uptake and survival among metastatic breast cancer patients in Lagos, Nigeria
The intersection of depression and cancer outcomes represents a critical yet understudied domain in global oncology, particularly in low- and middle-income countries (LMICs) where both conditions impose substantial disease burdens. While evidence from high-income countries demonstrates associations between depression and adverse cancer outcomes, the generalizability of these findings to resource-limited settings remains uncertain. Theoretical frameworks including Beck’s cognitive model and the biopsychosocial model suggest multiple pathways through which depression may influence cancer outcomes; however, these pathways may operate differently in resource-constrained healthcare environments. This study examines the relationship between clinically significant depression (CSD) and treatment uptake and survival among metastatic breast cancer patients in Nigeria, providing novel insights into psycho-oncology within the sub-Saharan African context. We conducted a cross-sectional analysis of 313 metastatic breast cancer patients presenting to NSIA-LUTH Cancer Centre, Lagos, Nigeria (September 2020–February 2022). Depression was assessed using the Beck Depression Inventory-II (BDI-II), with CSD defined as BDI-II ≥ 20, consistent with oncology-validated cutoffs and African psychometric validation studies. Associations between CSD and treatment receipt were evaluated using multivariable logistic regression. Survival patterns were examined using Kaplan-Meier analysis and Cox proportional hazards models, adjusting for clinicopathological confounders. The cross-sectional design precludes causal inference regarding temporal relationships between depression and outcomes. Among 274 patients with complete data (mean age 54.0 years, 52.1% triple-negative), depression was prevalent (52.6% any depression; 17.9% CSD). Treatment uptake was 64.6% overall. Contrary to findings from high-income settings, CSD showed no significant association with treatment receipt (adjusted OR 0.95, 95% CI 0.36–2.52, p = 0.92) or mortality (adjusted HR 1.12, 95% CI 0.48–2.61, p = 0.80). Clinical factors dominated outcome prediction: lung metastasis (OR 3.63), ulceration (OR 20.39), and lymph node status significantly predicted treatment receipt, while multimodal therapy was strongly protective against mortality (HR 0.25, 95% CI 0.10–0.65). Findings should be interpreted in the context of 12.5% missing depression data and potential selection and misclassification biases. In this Nigerian cohort, CSD was common but not associated with treatment uptake or survival. These findings suggest that in resource-constrained settings where clinical triage predominates, tumor biology and treatment access may supersede psychosocial factors as outcome determinants. The substantial depression prevalence nonetheless warrants integrated mental health services within cancer care, with implications for comprehensive oncology models in LMICs.
- Open Access
- 16 October 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (10) , e0311694
Development of a pharmacy-based HIV PrEP service delivery intervention for Washington, District of Columbia (DC): A study protocol
Pharmacy-based PrEP service delivery models can help address many of the barriers that inhibit the uptake of PrEP. In an increasing number of states, legislation has been passed, or is under consideration, to allow pharmacists to initiate PrEP without a prescription from a physician or other prescriber. However, there is not yet legislation in Washington, DC to allow pharmacy-based PrEP despite its potential to curb new cases of HIV, which disproportionately affect the Black community in the area. The DC Ends HIV Plan has a goal of less than 130 new cases of HIV per year by 2030, which would require that over 13,000 high-risk residents use PrEP. However, in 2021 only 6,724 Washingtonians were taking PrEP. This study seeks to address the absence of critical formative research into the factors that would influence the implementation of pharmacy-based PrEP in Washington DC using the Implementation Mapping (IM) framework. A needs assessment will be conducted through in-depth interviews (IDIs) with pharmacists (n = 6), PrEP providers (n = 6), current PrEP users (n = 6), DC Department of Health officials (n = 2), DC Board of Pharmacy officials (n = 4) and pharmacy-based PrEP experts (n = 4) to provide input on the operational aspects of pharmacy-based PrEP model as a strategy to increase PrEP uptake. Information gathered through this needs assessment will be used to develop standard operating procedures for the introduction of pilot pharmacy-based PrEP into community-based retail pharmacies.
- 16 October 2024
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (10) , e0311694
Open AccessDevelopment of a pharmacy-based HIV PrEP service delivery intervention for Washington, District of Columbia (DC): A study protocol
Pharmacy-based PrEP service delivery models can help address many of the barriers that inhibit the uptake of PrEP. In an increasing number of states, legislation has been passed, or is under consideration, to allow pharmacists to initiate PrEP without a prescription from a physician or other prescriber. However, there is not yet legislation in Washington, DC to allow pharmacy-based PrEP despite its potential to curb new cases of HIV, which disproportionately affect the Black community in the area. The DC Ends HIV Plan has a goal of less than 130 new cases of HIV per year by 2030, which would require that over 13,000 high-risk residents use PrEP. However, in 2021 only 6,724 Washingtonians were taking PrEP. This study seeks to address the absence of critical formative research into the factors that would influence the implementation of pharmacy-based PrEP in Washington DC using the Implementation Mapping (IM) framework. A needs assessment will be conducted through in-depth interviews (IDIs) with pharmacists (n = 6), PrEP providers (n = 6), current PrEP users (n = 6), DC Department of Health officials (n = 2), DC Board of Pharmacy officials (n = 4) and pharmacy-based PrEP experts (n = 4) to provide input on the operational aspects of pharmacy-based PrEP model as a strategy to increase PrEP uptake. Information gathered through this needs assessment will be used to develop standard operating procedures for the introduction of pilot pharmacy-based PrEP into community-based retail pharmacies.
- Open Access
- 1 January 2019
- Published by Medknow in Journal of Family Medicine and Primary Care
- Vol. 8 (12) , 3808-3813
The precedence and viability of yoga in the lives of D3-dental students, dental practitioners, and dental patients
Dentistry as a profession is quite demanding and challenging and is also associated with occupational hazards like musculoskeletal disorders and stress-related problems. The practice of yoga has been proven to be efficacious in management of quite a lot of stress and lifestyle disorders. It strengthens one physically, mentally, and emotionally. Moreover, yoga can also be used in the management of dental patients for stress-related oral conditions like lichen planus, burning mouth syndrome, aphthous ulcers, etc., Yoga has also been found to have antiinflammatory, antianxiety effect, and also a lot of other beneficial effects. This review emphasizes on the relevance of yoga in the dental profession, from the practitioners, students, and patients viewpoint. Data pertaining to the subject was obtained after a thorough search of the PubMed literature and Google search engine.
- 1 January 2019
- Published by Medknow in Journal of Family Medicine and Primary Care
- Vol. 8 (12) , 3808-3813
Open AccessThe precedence and viability of yoga in the lives of D3-dental students, dental practitioners, and dental patients
Dentistry as a profession is quite demanding and challenging and is also associated with occupational hazards like musculoskeletal disorders and stress-related problems. The practice of yoga has been proven to be efficacious in management of quite a lot of stress and lifestyle disorders. It strengthens one physically, mentally, and emotionally. Moreover, yoga can also be used in the management of dental patients for stress-related oral conditions like lichen planus, burning mouth syndrome, aphthous ulcers, etc., Yoga has also been found to have antiinflammatory, antianxiety effect, and also a lot of other beneficial effects. This review emphasizes on the relevance of yoga in the dental profession, from the practitioners, students, and patients viewpoint. Data pertaining to the subject was obtained after a thorough search of the PubMed literature and Google search engine.