Skip to content

Latest commit

 

History

History
81 lines (45 loc) · 2.57 KB

README.md

File metadata and controls

81 lines (45 loc) · 2.57 KB
Raw

maxHR

maximum heart rate estimation from noisy data

Yousof Erfani : 19 June 2017

The code to compute the (HRmax) max heart rate: Below is my analysis for my approach:

  1. As the data is noisy, I first estimate a polynomial fit (or any other better fitting method) to the data, in this case the effect of the noise of recording is reduced (The blue lines).

  2. Also, the heart rate signal is not stationary so the heart rate variability is changing during the time. So, I estimated the short term variance of hr signals for short intervals (npadd variable , defualt = 100, still to play with).

In this case, if there are a lot of variation on the signal in a short time, I ignore those segments with high short time variance when computing the hrMax.

  1. I computed the polynomial fitted values + 3* short time standard deviation for 2* npad time samples around each time sample
    I have plotted this in magenta color.

  2. Max of the plot in magenta color is the estimated hrMax if the variance smaller than a threshold (here Th =5 ), The place of the estimation for hrMAx is shown by a black circle.

The value of the maxHR estimation is shown in the command window after running the function. For each runner the estimate of the maxHR can be the average of all estimations.

If I had more time: 1 would try to remove outliers using more domain knowledge. I will denoise the hrSignals more using digital filtering

I will search more on the internet for efficient hr variability measuring from the data.
I will play more with the parameters I introduced to find the best estimate of the maxHR

--------- A sample results npolyfit = 50 Th_variance = 5

Maximum HR for user1Run1.csv is 176.7

my estimatred Maximum HR for user1Run1.csv is 172.5

Maximum HR measured at index: 3591.0

my estimatred Maximum HR for user1Run2.csv is 163.5

Maximum HR measured at index: 1575.0

Maximum HR for user1Run3.csv is 173.0

my estimatred Maximum HR for user1Run3.csv is 168.7

Maximum HR measured at index: 4099.0

Maximum HR for user2Run1.csv is 140.4

my estimatred Maximum HR for user2Run1.csv is 137.0

Maximum HR measured at index: 4484.0

Maximum HR for user2Run2.csv is 154.2

my estimatred Maximum HR for user2Run2.csv is 150.2

Maximum HR measured at index: 6364.0

Maximum HR for user2Run3.csv is 160.0

my estimatred Maximum HR for user2Run3.csv is 159.3

Maximum HR measured at index: 8668.0