Malaria remains a major public health challenge in sub-Saharan Africa, particularly in Nigeria, where delayed diagnosis contributes significantly to morbidity and mortality. Conventional diagnostic approaches, including microscopy and rapid diagnostic tests, are often limited by inadequate healthcare infrastructure, shortage of skilled personnel, and inconsistent diagnostic accuracy in underserved communities. This study proposes a multimodal explainable artificial intelligence framework for early malaria detection in resource-constrained Nigerian healthcare settings. The framework integrates clinical symptoms, hematological parameters, and microscopic blood smear images to improve diagnostic performance and reliability. Structured clinical and laboratory data are analyzed using ensemble machine learning techniques, while convolutional neural networks are employed for image-based parasite detection. A multimodal fusion strategy is adopted to combine complementary information from the different data sources for improved predictive accuracy. To enhance transparency and clinical trust, the framework incorporates explainable artificial intelligence techniques, including SHapley Additive exPlanations (SHAP), Local Interpretable Model-Agnostic Explanations (LIME), and Gradient-weighted Class Activation Mapping (Grad-CAM). These methods provide interpretable insights into model predictions and support visual identification of infected regions in blood smear images. The developed system is further designed for scalability and deployment on lightweight mobile and edge-based healthcare platforms suitable for low-resource environments. The study conceptually demonstrates that combining multimodal learning with explainable artificial intelligence can improve malaria diagnosis, strengthen clinical decision-making, and support accessible healthcare delivery in underserved communities.