Photoplethysmogram Signal Conditioning by Monitoring of Oxygen Saturation and Diagnostic of Cardiovascular Diseases

High performance biosignal preparation for the noninvasive diagnosis of diseases or calculation of physiological parameters is a very important task. High resolution and high signal noise ratio have to be achieved by the hardware design using a high resolution ADC and a PGA with the desired filters. After the detection of the signals by the calculation units (MC, DSP and PC) the signal conditioning will be continued. When the signal is measured, it is subjected to noise of different sources that has to be canceled at the first step. The motion artifacts by photoplethysmogm (PPG) are very difficult to eliminate with conventional filters. The frequency of the resulting noise signal has the same range of that of the PPG signal. Therefore, we use an adaptive filter using least mean square method (LMS) for this purpose. The second stage is the extraction of the distinctive features or characteristics of the filtered signal. One of the parameters values is the detection of peaks and valleys of the PPG signal. These parameters are used for the calculation of the heart rate, oxygen saturation and the hemoglobin concentration. Also other features like notch, the rise time from valley to peak and form of PPG have to be classified for the diagnostic of cardiovascular diseases. This paper describes an adaptive filter and a reliable peak and valley detection method used by the multisensor developed for monitoring of vital parameters and diagnosis of cardiovascular diseases. The proposed algorithms were applied to different PPG signals as will be discussed in the results.