Patch-Transformer Network: A Wearable-Sensor-Based Fall Detection Method

Emerging Digital Technologies Used for Fall Detection in Older Adults in Aged Care: A Scoping Review

OBJECTIVE

To explore a comprehensive overview of digital technologies used for fall detection in older adults, categorizing the types, functions, and usability of these systems.

DESIGN

A scoping review was conducted to search across 5 databases [Embase, Medline (OVID), CINAHL, Coherence and IEEE Explore] from January 2013 to September 2023.

SETTING AND PARTICIPANTS

Studies in older adults living in nursing homes, care homes, residential homes, respite care homes, and all skilled and ambulatory care facilities (without context restrictions).

METHODS

This review followed the 6 methodological stages: (1) identification of research question; (2) identification of relevant studies; (3) study selection; (4) charting the data; (5) collating, summarizing, and reporting the results; and an optional stage, (6) consulting with stakeholders regarding findings to explore pivotal concepts in emerging technology usage in long-term care for falls detection among older people. Data were extracted and categorized based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines.

RESULTS

A total of 73 studies met the inclusion criteria. Four main categories of fall detection technologies were identified: motion and sensor technologies, imaging and visual systems, environmental sensors, and robotic and autonomous systems. Commonly used devices: wearable accelerometers, gyroscopes, infrared array sensors, and smart carpet pressure sensors. Data storage methods were wearable devices, cameras, and floor-mounted sensors. Communication technologies included Bluetooth, Wi-Fi, and GPS, and notification methods ranged from alarms and SMS to cloud communications. Various health care response teams, including caregivers, health care providers, and emergency services, were integral to the fall detection systems.

CONCLUSIONS AND IMPLICATIONS

Most studies primarily focus on fall detection; however, we recommend further clinical research to emphasize both fall detection and, more importantly, fall prevention (both primary and secondary). Investigating the effectiveness of fall prevention technologies in real-world settings will be crucial for enhancing the safety and quality of life of the aging population.

Experimental Study of Long Short-Term Memory and Transformer Models for Fall Detection on Smartwatches

Falls are the second leading cause of unintentional injury deaths worldwide. While numerous wearable fall detection devices incorporating AI models have been developed, none of them are used successfully in a fall detection application running on commodity-based smartwatches in real time. The system misses some falls, and generates an annoying amount of False Positives for practical use. We have investigated and experimented with an LSTM model for fall detection on a smartwatch. Even though the LSTM model has high accuracy during offline testing, the good performance of offline LSTM models cannot be translated to the equivalence of real-time performance. Transformers, on the other hand, can learn long-sequence data and patterns intrinsic to the data due to their self-attention mechanism. This paper compares three variants of LSTM and two variants of Transformer models for learning fall patterns. We trained all models using fall and activity data from three datasets, and the real-time testing of the model was performed using the SmartFall App. Our findings showed that in the offline training, the CNN-LSTM model was better than the Transformer model for all the datasets. However, the Transformer is a preferable choice for deployment in real-time fall detection applications.

Fall Detection System Based on Point Cloud Enhancement Model for 24 GHz FMCW Radar

Automatic fall detection plays a significant role in monitoring the health of senior citizens. In particular, millimeter-wave radar sensors are relevant for human pose recognition in an indoor environment due to their advantages of privacy protection, low hardware cost, and wide range of working conditions. However, low-quality point clouds from 4D radar diminish the reliability of fall detection. To improve the detection accuracy, conventional methods utilize more costly hardware. In this study, we propose a model that can provide high-quality three-dimensional point cloud images of the human body at a low cost. To improve the accuracy and effectiveness of fall detection, a system that extracts distribution features through small radar antenna arrays is developed. The proposed system achieved 99.1% and 98.9% accuracy on test datasets pertaining to new subjects and new environments, respectively.