All posts by Miguel Ramirez

July 2-6

It was programmed a second part of GUI. This part displays and analyse the bi-harmonic pulse, its first and second derivatives. It displays also important points on these curves, like the systolic and diastolic peaks, dicrotic notch, and other points on the first and second derivatives useful to determine vascular information.

 

The computed values displayed in this analysis results window are

-Systolic and Diastolic Peaks            -Systolic-Diastolic Interval

-Systolic Amplitude                                -Augmentation Index

-Dicrotic Notch                                         -Pulse Interval

-Crest Time                                                  -Vascular Age

-Arterial Stiffness Index                        -Vascular Aging Index

-Peripheral Resistance                           -APG Waveform Types

The next steps are:

-Improve GUI data reading and first data analysis capabilities. Improve algorithm to calculate vascular age.  Create more efficient optimization algorithms to determine critical points.  Include more harmonics in the analysis.

 

Jun 25 -29

During this week:

  • A data base of Photoplethysmography data containing 42 cases, 8 minutes long, (29 children and 13 adults) was downloaded and processed using MathLab and Microsoft Excel. The data files will be saved in a the final csv format. Some of this data, even filtered and processed, present some kind of DC noise, but are still very useful for this research

(Data source: http://www.capnobase.org/database/pulse-oximeter-ieee-tbme-benchmark/)

  • The GUI interface was tested with some of this data. A good match to the data in the times was achieved even with only the first two harmonics of the Fourier series. This analytical expressions can be considered a filtered signal.

There is a limitation in this graphical approach: The precisions of the sliders. By now is only possible a two decimals precision in the values.

  • Some research works have been explored to understand how to process this analytical function fitted to the PPG data in order to obtain information about the cardiovascular condition of the these persons. If was found that the first and second derivatives can give numbers that correlate with artery stiffness, a problem that comes with aging. Numbers obtained from these derivatives may help to diagnose other cardiovascular diseases like hypertension.
  • The possibility to integrate in the GUI a routine to calculate the derivatives of the analytical expressions and obtain the necessary information from them was considered.
  • First draft of Research summary was written.

Jun-25

Meeting with project mentors. Presentation of GUI

Version 1.0  of Graphic User Interface ready. Written in Python 2.7.10

GUI PPG data – Harmonic Function-1ko0fme

Capabilities:

  • Reads PPG data from a csv file
  • Analyze data to determine: data average amplitude, data average vertical shift, maximums and minimums, and average frequency of the data
  • Generates an harmonic function using a linear combination of Sine and Cosine:                                                      f(t) = A sin(2*pi*w1 +c) + B cos(2*pi*w2+d)
  • Makes a first fit of the harmonic function to the data
  • Displays a graph with the PPG data and the fitted harmonic function
  • Sliders on the graphing window allow to modify the amplitudes  A, B, the frequencies w1, w2, the horizontal shifts c, d, to have a better fit of the PPG data points and the harmonic function

This GUI produces  a good  match for low noise PPG data