Authors: Sawant Pranjal Parshuram, Dr. Khatri A.A, Professor Bhosale S.B, Professor Banger A.P
Abstract: The growing number of older adults has increased the need for monitoring tools that can support safety and help care-givers respond quickly. Age-related changes in balance, strength, and movement may make falls more likely, while a person who lives alone may not be able to call for help immediately. This review looks at three studies that use different ways to identify falls. Care Fall analyses smartwatch accelerometer and gyroscope signals with threshold rules and ma-chine-learning methods. A second system combines motion and basic health sensors with a microcontroller and GSM/Wi-Fi alerts. The third approach uses a camera, the YOLOv5s model, and an edge-computing device for visual detection. Together, the studies show that wearable sensors, embedded systems, and camera-based AI can all support fall monitoring, but they differ in cost, privacy, response time, and practical use. Common concerns include false alarms, changes in fall patterns, environmental limitations, battery life, and data protection. The review compares the systems and outlines areas that need further testing before these approaches can be relied on in everyday care.
