As the number of elderly people increases, the demand of vital sign continuous measurement devices for health monitoring is increasing. We are aiming at the development of wearable vital signs measuring equipment using Fiber Bragg Grating (FBG) sensor which is an optical fiber type strain sensor. In this report, blood pressure prediction was performed on seven elderly people with different pulse wave patterns, and the results were examined. The blood pressure was predicted by constructing a calibration model by Partial Least Squares Regression (PLSR) from the reference blood pressure measured simultaneously with the pulse wave measured by the FBG sensor. As a result, the blood pressure prediction error of all subjects become 6mmHg for systolic blood pressure and 4mmHg for diastolic blood pressure, which indicate the possibility of blood pressure prediction using FBG sensor in elderly people with different pulse wave patterns. In addition, it is possible to construct a calibration model for more accurate blood pressure prediction.
