Name
Abstract | ||
|---|---|---|
The high blood pressure is an important risk factor for cardiac diseases. The blood pressure changes easily due to both physical and mental states. Therefore, the continuous blood pressure measurement device without physical stress has been demanded. However, it is difficult for the conventional measuring devices to measure continuously the blood pressure without consciousness. In order to respond to these problems, we have proposed using the Fiber Bragg Grating sensor to develop a blood pressure measurement device which can measure continuously, non-invasive and unconstrained. However, the prediction method which meets the demanded accuracy only when it is personalized to the individual. In this paper, we compared the prediction accuracy of two method - Partial Least Squared Regression (PLSR) and Artificial Neural Network (ANN). We confirmed the individual difference of the pulse waveform affected the prediction accuracy. The effect was able to be reduced by the repetitive learning of ANN. Consequently, ANN is the appropriate method for the blood pressure prediction toward to develop the versatile device. |
Year | Venue | Field |
|---|---|---|
2018 | MeMeA | Measuring instrument,Fiber Bragg grating,Regression,Computer science,Waveform,Partial least squares regression,Pulse (signal processing),Blood pressure,Acoustics,Artificial neural network |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Kyoko Katayama | 1 | 0 | 0.34 |
| hiroaki ishizawa | 2 | 5 | 3.89 |
| Shouhei Koyama | 3 | 4 | 3.46 |
| K Fujimoto | 4 | 2 | 2.31 |