摘要:
合集:AI案例-ML-人类生理心理
数据集:通过手杖检测出的心率等生理参数数据集
数据集价值:通过手杖快速识别跌倒事件
一、问题描述
尽管市场上有很多针对老年人的可穿戴设备提供关怀,但它们大多只关注跌倒的检测而不是预测。一些提出的可穿戴设备能够检测到绊倒和滑倒,但并不专门针对跌倒。因此,提出了一种名为cStick的手杖来监测老年人的跌倒情况。cStick的设计旨在帮助视觉和听力受损的老年人。cStick不仅能检测跌倒,还能预测跌倒事件的发生,从而减少跌倒的发生。cStick可以监控周围环境,如果之前在某个位置检测到跌倒,就会警告用户,并向用户更新该位置及其周围环境。根据监测参数的变化,做出跌倒决策,即预测、警告或检测跌倒,准确度约为95%。
cStick设备组件图:

二、数据集内容
数据结构
cStick.csv文件包含了大约2,040个数据样本,参数包括——距离、压力(0-小压力,1-中等压力,2-高压力)、HRV(心率变异性)、血糖水平、血氧饱和度水平和加速度计读数(<+-3g,即阈值为0,>阈值为1),以及与跌倒决策(0-未检测到跌倒,1-人绊倒/滑倒/预测跌倒,2-确定跌倒)的关系。
字段定义如下:
Distance:距离
Pressure:压力(0-小压力,1-中等压力,2-高压力)
HRV:心率变异性
Sugar level:血糖水平
SpO2:血氧饱和度水平
Accelerometer:加速度计读数(<+-3g,即阈值为0,>阈值为1)
Decision:是否跌倒的决策(0-未检测到跌倒,1-人绊倒/滑倒/预测跌倒,2-确定跌倒)
样例:cStick.csv
Distance | Pressure | HRV | Sugar level | SpO2 | Accelerometer | Decision |
---|---|---|---|---|---|---|
25.54 | 1 | 101.396 | 61.08 | 87.77 | 1 | 1 |
2.595 | 2 | 110.19 | 20.207 | 65.19 | 1 | 2 |
68.067 | 0 | 87.412 | 79.345 | 99.345 | 0 | 0 |
13.09 | 1 | 92.266 | 36.18 | 81.545 | 1 | 1 |
69.43 | 0 | 89.48 | 80 | 99.99 | 0 | 0 |
27.16 | 1 | 102.584 | 64.32 | 88.58 | 1 | 1 |
57.134 | 0 | 70.824 | 73.69 | 93.69 | 0 | 0 |
66.356 | 0 | 84.816 | 78.46 | 98.46 | 0 | 0 |
60.382 | 0 | 75.752 | 75.37 | 95.37 | 0 | 0 |
23.17 | 1 | 99.658 | 56.34 | 86.585 | 1 | 1 |
数据集引用要求
If this research or the dataset provided help you in anyway, please cite:
L. Rachakonda, A. Sharma, S. P. Mohanty, and E. Kougianos, "Good-Eye: A Combined Computer-Vision and Physiological-Sensor based Device for Full-Proof Prediction and Detection of Fall of Adults", in Proceedings of the 2nd IFIP International Internet of Things (IoT) Conference (IFIP-IoT), 2019, pp. 273--288.
L. Rachakonda, S. P. Mohanty, and E. Kougianos, “Good-Eye: A Device for Automatic Prediction and Detection of Elderly Falls in Smart Homes”, in Proceedings of the 6th IEEE International Symposium on Smart Electronic Systems (iSES), 2020, pp. 202--203.
L. Rachakonda, S. P. Mohanty, and E. Kougianos, “cStick: A Calm Stick for Fall Prediction, Detection and Control in the IoMT Framework”, in Proceedings of the 4th IFIP International Internet of Things (IoT) Conference (IFIP-IoT), 2021, pp. Accepted on 02 Sep 2021.
三、机器学习样例
这是传统机器学习二元分类案例,源码:sean-sayed-cstick-fall-prediction.ipynb。