Explainable Artificial Intelligence for Medical Data Analytics and Healthcare Applications

E-healthcare system applications utlising the Internet of Things (IoT) have revolutionized the medical system for patient monitoring, continuous real-time diagnostics, and advanced clinical communication. However, these solutions are dependent on the medicalcondition being considered for diagnosis. To the best of your knowledge, E-Healthcare systems utilizing IoT for managing arterial hypertension during pregnancy have been identified as a higher risk to both maternal and fetal health. In this chapter, we present the Explainable Artificial Intelligence (XAI)based Dashboard that provides E-Healthcare system analytics using XAI, specifically designed and developed for the early recognition and supervision of hypertension during pregnancy. The designed system assimilates a variety of different data, including physical parameters such as pulse, blood oxygen, ECG, and others. All these parameters are preprocessed and given to a hybrid machine learning model that comprises neural networks and an Ensemble XGBoost, which is highly trained on simulated data from 15,420 clinical samples. XAI enhances reliability in clinical decision-making through SHAP (SHapley Additive exPlanations). The designed system enables users to visualize and understand how individual features impact each prediction. XAI introduces trust by providing transparent insights into addressing multiple risk factors, such as BMI, gestational age, and a prior history of hypertension. This will help clinicians make well-informed and reasonably aware system decisions and support confidence in the AI system’s use.. The real-time XAI insights, provided through dashboards, enhance the goal of trustworthy and reliable maternal healthcare powered by AI.