This paper focuses on the design and foundational implementation of a Centralized Emergency Room (ER) Monitoring System, an advanced Internet of Medical Things (IoMT) solution aimed at revolutionizing patient surveillance. The core objective is to achieve the real-time, wireless acquisition and centralized management of critical patient vital signs: Heart Rate (HR), Blood Oxygen Saturation (SpO2), and Body Temperature. The system utilizes a powerful two-tier architecture. The device layer is anchored by the ESP32 Development Board, selected for its dual-core processing and integrated Wi Fi/Bluetooth Low Energy (BLE) connectivity, which ensures robust, concurrent handling of sensor data and reliable transmission via HTTP/JSON. The gateway layer is a scalable Python Flask server that manages multi-patient data persistence in a SQLite database. Crucially, this server acts as the primary platform for the AI Anomaly Detection engine, which analyses incoming vital sign streams for subtle, non-threshold-based patterns indicative of patient deterioration, thereby offering a superior level of risk assessment than traditional systems.