PRISM: Privacy Preserving Internet of Things Security Management
Consumer healthcare applications are gaining increasing popularity in our homes and hospitals. These devices provide continuous monitoring at a low cost and can be used to augment high-precision medical devices. The majority of these devices aim to provide anomaly detection for the users while aiming to provide data privacy and security by design using edge computing. However, major challenges remain in applying pre-trained global models on smart health monitoring devices, for a diverse set of individuals that they provide care for. In this paper, we propose PRISM, an edge-based system for in-home smart healthcare anomaly detection. We develop and release a rigorous methodology that relies on automated IoT anomaly experimentation. We use a rich, highly granular real-world dataset from in-home patient monitoring from 44 homes of people living with dementia (PLWD) over two years. Our results indicate that the anomalies can be identified with accuracy up to 99 as low as 0.88 seconds. While all models achieve high accuracy when trained on the same patient, their accuracy degrades when evaluated on different patients.
READ FULL TEXT