Edge Artificial Intelligence: Real-Time Noninvasive Technique for Vital Signs of Myocardial Infarction Recognition Using Jetson Nano

Table 1

Fall detection techniques.

Author	Datasets	No. of subjects (age)	Sensor	Algorithms
Saleh and Jeannès [28]	Simulated	23 (19–30), 15 (60–75)	Accelerometer (waist)	SVM
Zitouni et al. [29]	Simulated	6 (N/A)	Accelerometer (sole)	Threshold
Wu et al. [30]	Public (simulated)	42 (N/A), 36 (N/A)	Accelerometer (chest and thigh)	Decision tree
Huang et al. [31]	Simulated	12 (19–29)	Vibration	HMM
Tian et al. [32]	Simulated	140 (N/A)	FMCW radio	CNN
Wang et al. [33]	Simulated	N/A	WiFi	SVM, Random Forests
Kerdjidj et al. [34]	Simulated	17 (N/A)	Accelerometer, gyroscope	Compressive sensing
Queralta et al. [35]	Public (simulated)	57 (20–47)	Accelerometer, gyroscope, magnetometer	LTSM
Han et al. [36]	Simulated	N/A	Web camera	CNN
Kong et al. [37]	Public	Public	Camera (surveillance)	CNN
Ko et al. [38]	Simulated	N/A	Camera (smartphone)	Rao-Blackwellized particle filtering
Shojaei-Hashemi et al. [39]	Public (simulated)	40 (10–15)	Kinect	LSTM
Min et al. [40]	Public (simulated)	4 (N/A), 11 (22–39)	Kinect	SVM
Ozcan et al. [41]	Simulated	10 (24–31)	Web camera	Relative-entropy-based