Averaging data from seconds to three seconds because of different time variables

I have measured data with two different devices.

One device (Garmin device)measured data every second for about 20 minutes, so I have data with a 1207x1 format.

The other device (Cosmed device) measured data every 3 seconds (so 1, 4, 7, 10, etc.) also for about 20 minutes, so this results in a 470x1 format.

I want to compare variables measured from these devices, but for a reliable analysis I have to have the same time dimension (either per 1 or 3 seconds).

The most simple option would be per 3 seconds, but in order to do that I have to ''average'' the data from the Garmin for every 3 seconds.

What is the simplest way to achieve this?

Thanks in advance!

[claudio]

you can use interp1 function interp1

% to adapt cosmed to garmin measurement
newDataCosmed = interp1(timeCosmed,dataCosmed,timeGarmin,'pchip');
% to adapt gamin to cosmed measurement
newDataGarmin = interp1(timeGarmin,dataGarmin,timeCosmed,'pchip');

[Dyuman Joshi]

Your data does not match.

1207 elements for data every second means, the measurements have been taken for 1207 seconds.

Then the number of elements for data measured every 3 seconds would be

numel(1:3:1207)

ans = 403

But that's not the case.

You will have to either trunk the data from Cosmed device (and Garmin device as 1207 is not divisble by 3) or extend the data from the Garmin device (to 470*3 measurement values), so that the data matches.

[Stephen23]

"You will have to either trunk the data from Cosmed device (and Garmin device as 1207 is not divisble by 3) or extend the data from the Garmin device (to 470*3 measurement values), so that the data matches."

It is very unlikely that two independent devices have saved data at synchronized time points, so simply truncating or extending is unlikely to to be sufficient.

More likely RETIME and/or SYNCHRONIZE would be required, as Star Strider suggests.

[Ilse Frans]

But what would be the most reliable option, extending the Cosmed data or shortening the Garmin data?

And how would I use RETIME and/or SYNCHRONIZE in the best way?

[Star Strider]

It would help to have at least a sample of both files.

Use the zip function to include both of them (since the files themselves may not be permitted because of their extensions) and then upload the .zip file.

[Ilse Frans]

Here you go, but do know that I already transferred the timestamps into seconds in my script. If you want I can also upload that one.

[Ilse Frans]

Just in case :)

[Stephen23]

"but do know that I already transferred the timestamps into seconds in my script."

Please provide the original, complete timestamps.

Answer 1

[Torsten]

I'd repeat the data for 3 seconds three times. This does not manipulate your data with interpolation or averaging.

[Ilse Frans]

You mean that I wouldn't need the functions ''retime'' and/or ''synchronize'' then?

And (I'm sorry, I'm still new at using Matlab) how would I write the code to achieve that?

Answer 2

[Star Strider]

I am not certain that it is possible, however if it is, some combination of retime and synchronize could work.

EDIT — (21 Aug 2023 at 12:41)

Please provide the complete ‘garmin_timestamp’ with all the fields. Only having the seconds values is not going to work.

The ‘cosmed’ times appear to have only the minutes and seconds. It would help to have the hours aa well, if that is possible. (We can assume for the time being that they are the same days.)

type('garmin_cosmed_marc.m')

clear all;
close all;

garmin_alldata = readtable('Pilot_Marc Garmin.csv');
cosmed_alldata = readtable('Cosmed_Marc 27 juni.csv');

% [Belangrijke informatie]
% De huidige data is gemeten bij een proefpersoon met de volgende gegevens:
gender = 0; % Waarbij 0 = M, 1 = V
height = 1.85; % in meter
weight = 103; % in kg
age = 46; % in jaar, geboortedatum = 31/03/1977

% [Garmin variabelen]
garmin_cadance = garmin_alldata{:,2}; % cadans in aantal stappen/minuut
garmin_distance = garmin_alldata{:,3}; % afstand in meter
garmin_speed = garmin_alldata{:,4}; % snelheid in meter/seconde
garmin_HR = garmin_alldata{:,6}; % in bpm
garmin_power = garmin_alldata{:,7}; % vermogen in Watt
garmin_steplength_mm = garmin_alldata{:,11}; % staplengte in mm
garmin_steplength = garmin_steplength_mm/1000; % staplengte in m
garmin_timestamp = garmin_alldata{:,12}; % tijdsverloop in yyyy:mm:dd hh:mm:ss
garmin_t = time2num(garmin_timestamp);
garmin_oscillation_cm = garmin_alldata{:,14}; % verticale oscillatie in cm
garmin_oscillation = garmin_oscillation_cm/100; % verticale oscillatie in m
garmin_resprate = garmin_alldata{:,17};
garmin_resprate_real = garmin_resprate/1000; % respiration rate in brpm

% [Cosmed variabelen]
cosmed_t = cosmed_alldata{:,10}; % tijdsduur in seconden
cosmed_t = cosmed_t(4:end);
% [cosmed_t van mm:ss omzetten naar seconden en double value]
% Initialize an array to store the converted values
cosmed_seconds = zeros(size(cosmed_t));
% Iterate over each timestamp and convert it to seconds
for i = 1:numel(cosmed_t)
    % Split the timestamp string into minutes and seconds
    time_parts = split(cosmed_t{i}, ':');
    
    % Convert minutes and seconds to double values
    minutes = str2double(time_parts{1});
    seconds = str2double(time_parts{2});
    
    % Calculate the total time in seconds
    total_seconds = minutes * 60 + seconds;
    
    % Store the result in the array
    cosmed_seconds(i) = total_seconds;
end

cosmed_Rf = cosmed_alldata{:,11}; % ademhalingsfrequentie per minuut
cosmed_VE = cosmed_alldata{:,13}; % ademminuutvolume in L/min
cosmed_VO2 = cosmed_alldata{:,15}; % zuurstofopname in mL/min
cosmed_VCO2 = cosmed_alldata{:,16}; % kooldioxide uitademing in mL/min
cosmed_RQ = cosmed_alldata{:,17}; % respiratie quotiënt / RER
cosmed_VO2kg = cosmed_alldata{:,22}; % zuurstofopname in mL/min per kg lichaamsgewicht
cosmed_METs = cosmed_alldata{:,23}; % metabole quotiënt
cosmed_HR = cosmed_alldata{:,24}; % hartslag in bpm
cosmed_cadance = cosmed_alldata{:,77}; % cadans in stappen/min
cosmed_HRR = cosmed_alldata{:,82}; % heart rate reserve in bpm
cosmed_EEm = cosmed_alldata{:,45}; % energieverbruik in kcal/min

cosmed_prot_day = cosmed_alldata{:,48}; % eiwitverbruik in kcal/dag
cosmed_prot = cosmed_prot_day*0.00069; % eiwitverbruik in kcal/min
cosmed_prot_percent = cosmed_alldata{:,51}; % percentage eiwitten
cosmed_fat_day = cosmed_alldata{:,49}; % vetverbruk in kcal/dag
cosmed_fat = cosmed_fat_day*0.00069; % vetverbruik in kcal/min
cosmed_fat_percent = cosmed_alldata{:,52}; % percentage vetten
cosmed_CHO_day = cosmed_alldata{:,50}; % CHO-verbruik in kcal/dag
cosmed_CHO = cosmed_CHO_day*0.00069; % CHO-verbruik in kcal/min
cosmed_CHO_percent = cosmed_alldata{:,53}; % percentage CHO

% [EE_Garmin (in kcal/min) op basis van HR]
garmin_HR_min = min(garmin_HR);
garmin_HR_max = max(garmin_HR);
garmin_HRR = ((garmin_HR-garmin_HR_min)./(garmin_HR_max-garmin_HR))*100;
garmin_EE_HR = 1.044 + (0.025*garmin_HRR) + (1.0108*weight) + (0.001177*garmin_HRR*weight); %in kcal/min

% [EE_Garmin (in kcal/min) op basis van GPS, Pandolf-Santee]
carried_load = 0.9; % in kg
terrain_factor = 1; % op loopband
grade(:,1:240) = 0; 
grade(:,241:480) = 0;
grade(:,481:721) = 25;
grade(:,722:962) = 10;
grade(:,963:1207) = 0;
speed(:,1:240) = 1.39;
speed(:,241:480) = 0.83;
speed(:,481:721) = 0.83;
speed(:,722:962) = 1.67;
speed(:,963:1207) = 0.83;
garmin_EE_GPS_W = (1.5.*weight) + (2.0.*(weight+carried_load)).*((carried_load/weight).^2) + (terrain_factor.*(weight/carried_load)).*(1.5.*speed.^2+0.35.*speed.*grade);
garmin_EE_GPS_W = garmin_EE_GPS_W.';
garmin_EE_GPS = garmin_EE_GPS_W*0.01433075379765; % in kcal/min

% [Verhouding naar VO2 en MET]
garmin_VO2_GPS = garmin_EE_GPS*0.8; % GPS data, zuurstofverbruik in mL/min
garmin_MET_GPS = garmin_EE_GPS*199; % GPS data, MET-waarden

garmin_VO2_HR = garmin_EE_HR*0.8; % HR data, zuurstofverbruik in mL/min
garmin_MET_HR = garmin_EE_HR*199; % HR data, MET-waarden

% [Data ZIP]
zip('garmin_cosmed.zip',{'garmin_alldata.mat','cosmed_alldata.mat'})

% [Data exporteren naar CSV]
%for i = 1:3:length(garmin_t)
%     garmin_avg = mean(garmin_t(i:i+2));
% end

%writematrix([garmin_EE_GPS garmin_VO2_GPS garmin_MET_GPS garmin_EE_HR garmin_VO2_HR garmin_MET_HR cosmed_EEm cosmed_VO2 cosmed_METs],'garmin_cosmed_marc.csv')

Uz = unzip('garmin_cosmed.zip');
LDG = load(Uz{1});
garmin_alldata = LDG.garmin_alldata

garmin_alldata = 2031×5 table
    HR    cadence     distance     grade    t 
    __    _______    __________    _____    __

    91        0               0       0      0
    91        0               0       0      1
    93      115               0       0      2
    95      115               0       0      3
    95       60               0       0      4
    96       57      0.00050952       0      5
    97       60       0.0013359     NaN      6
    96       57       0.0021872     NaN      7
    95        0       0.0021872     NaN      8
    95        0       0.0021872     NaN      9
    96        0       0.0021872     NaN     10
    96        0       0.0021872     NaN     11
    96        0       0.0021872     NaN     12
    96        0       0.0021872     NaN     13
    96        0       0.0021872     NaN     14
    96        0       0.0021872     NaN     15

LDC = load(Uz{2});
cosmed_alldata = LDC.cosmed_alldata(4:end,10:end);
cosmed_alldata.t = datetime(cosmed_alldata.t,'InputFormat','mm:ss', 'Format','mm:ss')

cosmed_alldata = 793×57 table
      t       Rf       VT        VE       IV      VO2       VCO2      RQ     02exp    CO2exp    VE/VO2    VE/VCO2    VO2/kg    METs    FeO2     FeCO2    FetO2    FetCO2    FiO2    FiCO2    PeO2    PeCO2    PetO2    PetCO2      Fase        Marker      Amb T    Analyz. Press.    PB    EEkc    EEh    EEm     EEtot    EEkg    Pro    Vet     CHO    Pro%    Vet%    CHO%    npRQ     Ti      Te     Ttot    Ti/Ttot    VD/VT e    LogVE    t Rel    Fase tijd    VO2/Kg%Pred     BR     O2 delay    CO2 delay    RH sample    Stride Cadence    VT/Vi    PaCO2_e
    _____    _____    _____    ______    ____    ______    ______    ____    _____    ______    ______    _______    ______    ____    _____    _____    _____    ______    ____    _____    ____    _____    _____    ______    ________    __________    _____    ______________    __    ____    ___    ____    _____    ____    ___    ____    ___    ____    ____    ____    ____    ____    ____    ____    _______    _______    _____    _____    _________    ___________    ____    ________    _________    _________    ______________    _____    _______

    00:28       20    0.524     10.48     542     74.25    64.042    0.86    105.2      4.2     122.3      141.8      0.99     0.3     20.08     0.8     16.29     4.12     20.9    0.07     145       6       117       30      {'NONE'}    {0×0 char}    22.8           51          36    745     768     518     21.6    0.36     0      6.9     0      243     275     0      46.9    1.18    1.82       3     0.39       0.69       1.02      0      {'00:00'}         3           93      1267        1022          74               0          0.44       32   
    00:31    23.44    0.681    15.961     585    520.63    421.38    0.81    116.8       22      27.5         34      6.94       2     17.15    3.23     16.02     4.16     20.9    0.07     124      23       116       30      {'NONE'}    {0×0 char}    22.8           51          36    745     768    3591    149.6    2.49     0     47.9     0     2337    1255     0      65.1    1.07    1.49    2.56     0.42       0.18      1.203      3      {'00:03'}        19         89.4      1267        1022          74               0          0.64       32   
    00:34     16.9    0.729    12.321     662    377.55     304.8    0.81    126.7     22.1      29.5       36.5      5.03     1.4     17.38    3.03     15.81     4.29     20.9    0.07     125      22       114       31      {'NONE'}    {0×0 char}    22.8           51          36    745     768    2603    108.5    1.81     0     34.7     0     1712     891     0      65.8    1.75     1.8    3.55     0.49       0.25      1.091      3      {'00:06'}        13         91.8      1267        1022          74               0          0.42       33   
    00:38    14.96    0.855    12.793     968       400    339.99    0.85    147.7     27.8      29.4       34.5      5.33     1.5     17.27    3.25     15.75     4.38     20.9    0.07     125      23       114       32      {'NONE'}    {0×0 char}    22.8           51          36    745     768    2785      116    1.93     0     37.1     0     1426    1359     0      51.2    1.14    2.87    4.01     0.28       0.23      1.107      4      {'00:10'}        14         91.5      1267        1022          74               0          0.75       34   
    00:42    14.85    1.208    17.941    1122    521.37     447.3    0.86    211.7     36.9      32.4       37.8      6.95       2     17.52    3.05     16.02      4.3     20.9    0.07     126      22       116       31      {'NONE'}    {0×0 char}    22.8           51          36    745     768    3637    151.5    2.53     0     48.5     0     1763    1873     0      48.5    1.43    2.61    4.04     0.35       0.28      1.254      4      {'00:14'}        19         88.1      1267        1022          74               0          0.84       33   
    00:46    18.18    0.969    17.618     700    499.74    428.64    0.86    170.6     28.9      32.7       38.1      6.66     1.9     17.61    2.98     16.04     4.27     20.9    0.07     127      22       116       31      {'NONE'}    {0×0 char}    22.8           51          36    745     768    3486    145.2    2.42     1     46.5     0     1693    1793     0      48.6    1.26    2.04     3.3     0.38       0.28      1.246      4      {'00:18'}        18         88.3      1267        1022          74               0          0.77       33   
    00:49     19.8    0.764    15.129     515    449.28     371.6    0.83    133.5       23      30.6         37      5.99     1.7     17.47    3.01     15.75     4.36     20.9    0.07     126      22       114       31      {'NONE'}    {0×0 char}    22.8           51          36    745     768    3112    129.7    2.16     1     41.5     0     1836    1276     0        59    1.15    1.88    3.03     0.38       0.27       1.18      3      {'00:21'}        16           90      1267        1022          74               0          0.66       34   
    00:52    19.17    0.676    12.958     702    408.78     336.7    0.82    116.7     21.5      28.4       34.5      5.45     1.6     17.26    3.18     15.47     4.55     20.9    0.07     124      23       112       33      {'NONE'}    {0×0 char}    22.8           51          36    745     768    2829    117.9    1.96     1     37.7     0     1703    1126     0      60.2    1.34    1.79    3.13     0.43       0.24      1.113      3      {'00:24'}        15         91.4      1267        1022          74               0           0.5       35   
    00:54    23.53    0.828    19.482     712    692.13    564.49    0.82    139.2     29.3      25.8       31.6      9.23     2.6     16.81    3.54     15.29     4.64     20.9    0.07     121      26       110       33      {'NONE'}    {0×0 char}    22.8           51          36    745     768    4781    199.2    3.32     1     63.7     0     3009    1772     0      62.9    0.93    1.62    2.55     0.36        0.2       1.29      2      {'00:26'}        25         87.1      1267        1022          74             137          0.89       36   
    00:57    24.69    0.477    11.778     460    399.09    314.86    0.79     81.2     15.6      25.2       31.9      5.32     1.5     17.02    3.27     15.26     4.56     20.9    0.07     123      24       110       33      {'NONE'}    {0×0 char}    22.8           51          36    745     768    2740    114.2     1.9     1     36.5     0     1974     766     0        72    0.89    1.54    2.43     0.37       0.18      1.071      3      {'00:29'}        14         92.2      1267        1022          74               0          0.54       35   
    01:01    12.35     2.41    29.753    2141    1058.7    888.07    0.84    404.5     87.8      27.3       32.5     14.12       4     16.78    3.64     15.43     4.65     20.9    0.07     121      26       111       34      {'NONE'}    {0×0 char}    22.8           51          36    745     768    7353    306.4    5.11     1       98     0     4045    3307     0        55    1.99    2.87    4.86     0.41       0.23      1.474      4      {'00:33'}        38         80.3      1267        1022          74             156          1.21       36   
    01:03    32.61    0.463    15.098     856    392.69    334.39    0.85     82.8     12.6      32.6       38.3      5.24     1.5     17.88    2.72      16.3     3.97     20.9    0.07     129      20       118       29      {'NONE'}    {0×0 char}    22.8           51          36    745     768    2735      114     1.9     1     36.5     0     1386    1349     0      50.7    1.08    0.76    1.84     0.59       0.22      1.179      2      {'00:35'}        14           90      1267        1022          74              92          0.43       31   
    01:06     30.3     0.54    16.364      83    683.21    534.64    0.78     87.1     21.5      20.8       26.6      9.11     2.6     16.13    3.98     14.66     5.01     20.9    0.07     116      29       106       36      {'NONE'}    {0×0 char}    22.8           51          36    745     768    4684    195.2    3.25     2     62.5     0     3476    1207     0      74.2    0.36    1.62    1.98     0.18       0.12      1.214      3      {'00:38'}        24         89.1      1267        1022          74               0           1.5       38   
    01:09    18.13    0.523      9.48     547    383.88    286.67    0.75     85.3     19.3      21.4       28.6      5.12     1.5     16.31    3.69     14.45     5.03     20.9    0.07     118      27       104       36      {'NONE'}    {0×0 char}    22.8           51          36    745     768    2610    108.7    1.81     2     34.8     0     2256     354     0      86.4    1.46    1.85    3.31     0.44       0.17      0.977      3      {'00:41'}        14         93.7      1267        1022          74               0          0.36       38   
    01:11     22.3    0.809    18.045     905    709.94    539.05    0.76    132.9     29.5      23.2       30.6      9.47     2.7     16.43    3.65     14.98     4.76     20.9    0.07     118      26       108       34      {'NONE'}    {0×0 char}    22.8           51          36    745     768    4841    201.7    3.36     2     64.5     0     3977     864     0      82.2    1.12    1.57    2.69     0.42       0.19      1.256      2      {'00:43'}        25           88      1267        1022          74               0          0.72       36   
    01:14    20.41    0.672    13.714     757    506.87    394.28    0.78    112.1     23.6      24.2       31.2      6.76     1.9     16.68    3.51     15.15     4.65     20.9    0.07     120      25       109       34      {'NONE'}    {0×0 char}    22.8           52          36    745     768    3471    144.6    2.41     2     46.3     0     2631     840     0      75.8    0.92    2.02    2.94     0.31       0.19      1.137      3      {'00:46'}        18         90.9      1267        1022          74               0          0.73       36   

.

[Ilse Frans]

Unfortunately this is the only data I have received and it's not possible to change.

The Cosmed timestamp is given in mm:ss and the Garmin timestamp in yyyy:mm:dd hh:mm:ss.

But now that I have the Garmin data in seconds and Cosmed per 3, how will I be able to use the function retime and/or synchronize to work?

Answer 3

[MinJi KANG]

Let the Garmin device's data as x

for i=1:3:length(x)

avg = mean(x(i:i+2))

end

If you do that, you can get avg(470x1 format) data.

Hope you can solve it.

[Ilse Frans]

I understand this, but when I have a Garmin variable with a 1207x1 size, I get the following error:

for i = 1:3:length(garmin_t)

garmin_avg = mean(garmin_t(i:i+2));

end

>>

Index exceeds the number of array elements. Index must not exceed 1207.

Error in garmin_cosmed_marc (line 31)

garmin_avg = mean(garmin_t(i:i+2));

[Dyuman Joshi]

Yes, that because your data does not match in size. See my comment above.