@inproceedings{oai:ipsj.ixsq.nii.ac.jp:00231574,
 author = {Rakkrit, Duangsoithong and Kulika, Pahukarn and Dujdow, Buranapanichkit and Rakkrit, Duangsoithong and Kulika, Pahukarn and Dujdow, Buranapanichkit},
 book = {Proceedings of Asia Pacific Conference on Robot IoT System Development and Platform},
 month = {Dec},
 note = {Blood Pressure (BP) monitoring provides crucial information for individual healthcare conditions. Generally, BP measurement uses cuff-based instruments, however, it might be inconvenient for pain-sensitive patients or the elderly and it can also get germy, especially in public places. This paper presents the cuff-less blood pressure classification to improve ease of use and reduce potential physical effects on pain-sensitive patients. The electrocardiogram signal (ECG), the photoplethysmography signal (PPG), and the combination of ECG and PPG were used to create models for blood pressure classification. The traditional Neural Network trains and classifies blood pressure values, which divide blood pressure into four classes. This study discovered that the results when using the combination of PPG and ECG signals provided the highest accuracy. In the future work, deep learning will be analyzed and compared with the results of neural networks., Blood Pressure (BP) monitoring provides crucial information for individual healthcare conditions. Generally, BP measurement uses cuff-based instruments, however, it might be inconvenient for pain-sensitive patients or the elderly and it can also get germy, especially in public places. This paper presents the cuff-less blood pressure classification to improve ease of use and reduce potential physical effects on pain-sensitive patients. The electrocardiogram signal (ECG), the photoplethysmography signal (PPG), and the combination of ECG and PPG were used to create models for blood pressure classification. The traditional Neural Network trains and classifies blood pressure values, which divide blood pressure into four classes. This study discovered that the results when using the combination of PPG and ECG signals provided the highest accuracy. In the future work, deep learning will be analyzed and compared with the results of neural networks.},
 pages = {35--36},
 publisher = {情報処理学会},
 title = {Cuff-Less Blood Pressure Classification from ECG and PPG Signals using Deep Learning},
 volume = {2023},
 year = {2023}
}