Buddika, B. PJayasinghe N.Perera A.N.MWijethunga D.NWeerasinghe, MDunuwila, O2026-09-202026-05-22B. P. Buddika, J. N., P. A. N. M, D. N. Wijethunga, M. Weerasinghe and O. Dunuwila, "PregAssist: Pregnancy Support Mobile Application for Pregnant Mothers and Doctors," 2026 6th International Conference on Computer Communication and Artificial Intelligence (CCAI), Nanjing, China, 2026, pp. 1169-1174, doi: 10.1109/CCAI69603.2026.11641972.979-833158248-7https://rda.sliit.lk/handle/123456789/5275Pregnant care involves constant follow-ups, timely risk detection, and successful interactions between mothers and medics. This paper introduces PregAssist, a mobile integrated support system that is a synthesis of four products, including fetal health decision support, physical health risk prediction, AI-based mental health monitoring, and AR-driven emergency training. The fetal health module provides explainable and offline-capable cardiotocography (CTG) classification to assist clinical decision-making. XGBoost classifier with SHAP-based interpretability is used to analyze physical risks to generate individual recommendations and alerts. Mental health assessment integrates questionnaire-based indicators with CNN-driven facial emotion recognition, while federated learning preserves data privacy through on-device training. AR-based deterministic expert system offers protocol-adherent emergency simulations to improve preparedness activity when facing high-level of risk. Through the convergence of predictive analytics, explainable AI, privacy preserving learning, and training design, PregAssist assists in proactive maternal attention in urban and resource constrained environments via combining a hybrid mobile architecture.enAugmented RealityDecision Support SystemExplainable AIFederated LearningFetal Health ClassificationMaternal Mental HealthRule-Based Expert SystemConference PaperDOI: 10.1109/CCAI69603.2026.11641972