This chapter explores the design and implementation of Machine Learning-Based Intrusion Detection Systems (ML-IDS) specifically tailored for Healthcare Internet of Things (HIoT) environments. With the growing deployment of wearable sensors, smart medical devices, and remote monitoring tools in patient care, HIoT networks are increasingly vulnerable to cyber threats such as data breaches, denial-of-service attacks, and malicious intrusions. Traditional security mechanisms often fall short due to the sensitive nature of health data and the resource constraints of medical devices. This chapter analyzes how supervised, unsupervised, and deep learning techniques can enhance anomaly detection in HIoT, while addressing challenges related to data collection, class imbalance, and model training. It further discusses the integration of edge and fog computing for real-time detection, evaluates existing datasets, and highlights advances in federated learning and explainable AI.