An IoT-based framework for remote fall monitoring
| Autor: | Abbes Amira, Abdulah Jarouf, Faycal Bensaali, Ayman Al-Kababji, Lisan Shidqi, Hamza Djelouat |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning Wearable sensing device Computer science 0206 medical engineering Biomedical Engineering Mobile application Health Informatics 02 engineering and technology Feature extraction algorithm selection Signal Machine Learning (cs.LG) Computer Science - Networking and Internet Architecture 03 medical and health sciences 0302 clinical medicine CWT 3-axis accelerometer Continuous wavelet transform Euclidean vector Networking and Internet Architecture (cs.NI) Residential gateway business.industry Pattern recognition 020601 biomedical engineering Support vector machine Feature (computer vision) Signal Processing Artificial intelligence Focus (optics) F1 score business 030217 neurology & neurosurgery |
| Popis: | Fall detection is a serious healthcare issue that needs to be solved. Falling without quick medical intervention would lower the chances of survival for the elderly, especially if living alone. Hence, the need is there for developing fall detection algorithms with high accuracy. This paper presents a novel IoT-based system for fall detection that includes a sensing device transmitting data to a mobile application through a cloud-connected gateway device. Then, the focus is shifted to the algorithmic aspect where multiple features are extracted from 3-axis accelerometer data taken from existing datasets. The results emphasize on the significance of Continuous Wavelet Transform (CWT) as an influential feature for determining falls. CWT, Signal Energy (SE), Signal Magnitude Area (SMA), and Signal Vector Magnitude (SVM) features have shown promising classification results using K-Nearest Neighbors (KNN) and E-Nearest Neighbors (ENN). For all performance metrics (accuracy, recall, precision, specificity, and F1 Score), the achieved results are higher than 95% for a dataset of small size, while more than 98.47% score is achieved in the aforementioned criteria over the UniMiB-SHAR dataset by the same algorithms, where the classification time for a single test record is extremely efficient and is real-time 30 Pages, 9 figures, 9 tables. This is a the Accepted Manuscript version of the article published in Biomedical Signal Processing and Control (URL: https://doi.org/10.1016/j.bspc.2021.102532) |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|