血壓波形是重要的生理資訊，隱含著心臟血管功能的重要訊息。對於橈動脈血壓波形的檢測，醫院加護病房或手術房大都以侵入式為之，但這種作法對於患者有潛在風險，因此，本論文提出一種以非侵入式的方法，應用粒子群聚最佳化法之系統識別，以重建橈動脈血壓波形，主要目的是希望利用非侵入的方式，利用手指光學血液容積信號(Photoplethysmography, PPG)，配合粒子群聚最佳化法(Particle Swarm Optimization, PSO)及fuzzy C-means建立最佳波形轉換函數庫，使不同特徵的光學血液容積信號均有最佳轉移函數，以獲得精確的連續橈動脈血壓波形。在作法上我們採用轉移函數庫的構想，以事先的測試信號，針對不同特徵，建立專屬轉移函數。爾後的量測信號，只要比對轉移函數庫中的PPG信號，找出最佳歸類叢集(cluster)，即可以此叢集轉移函數做波形轉換，可快速及準確的獲得連續的橈動脈血壓波形。 Waveforms of blood pressure contain very important information of life. Although blood pressure can be continuously measured by an intra arterial catheter, this invasive method introduces risks to patients. Therefore, a noninvasive method in measuring blood pressure waveforms is proposed in this paper, based on which we can use the signals of fingertip photoplethysmogram to reconstruct radial pressure waveforms. Characteristics of various photoplethysmogram will be categorized into 3 clusters by using fuzzy C-mean clustering. A particle swarm optimization scheme is then established to search for an optimal transfer function model for estimating the radial pressure waveforms. When the PPG signals of various characteristics become available, an optimization scheme based on Particle Swarm Optimization (PSO) is proposed to derive a transfer function bank for reconstructing continuous radial pressure waveforms for other patients. The optimization scheme based on PSO has successfully applied to derive a set of transfer function banks with satisfactory performance in providing estimates for actual blood waveforms. Experiment results show that correlation ratio of the transformed waveforms can be as high as 0.89, much better than the results via the ARX technique.
