Please see the following reference for more detailed description of this pipeline. If you wish to use the code, please reference this work in any resulting publication:
- T. M. Urner et al."Normative cerebral microvascular blood flow waveform morphology assessed with Diffuse Correlation Spectroscopy," Biomed Opt Express, in press (2023)
Please also reference the below publication if using the script delineator.m:
- BN Li, MC Dong & MI Vai (2010) On an automatic delineator for arterial blood pressure waveforms Biomedical Signal Processing and Control 5(1) 76-81.
For R code to implement time-waited averaging as discussed in the paper, plase see:
- S. Halvorson, "SvenR," https://github.com/svenhalvorson/SvenR.
This code was tested in the following MATLAB environment:
-----------------------------------------------------------------------------------------------------
MATLAB Version: 9.14.0.2206163 (R2023a)
MATLAB License Number: 621625
Operating System: macOS Version: 13.2.1 Build: 22D68
Java Version: Java 1.8.0_202-b08 with Oracle Corporation Java HotSpot(TM) 64-Bit Server VM mixed mode
-----------------------------------------------------------------------------------------------------
MATLAB Version 9.14 (R2023a)
Curve Fitting Toolbox Version 3.9 (R2023a)
Image Processing Toolbox Version 11.7 (R2023a)
Medical Imaging Toolbox Version 1.1 (R2023a)
Parallel Computing Toolbox Version 7.8 (R2023a)
Signal Processing Toolbox Version 9.2 (R2023a)
Statistics and Machine Learning Toolbox Version 12.5 (R2023a)
Symbolic Math Toolbox
(HINT: the command ver in MATLAB will give you this info for your environment)
Please see the Example_Data example input dataset. The code assumes that data is saved with the format [Subject ID]--[varname].mat, as shown below. File handling within pulse_analysis_main.m is mostly in parse_db.m.
| Variable Name | Description | Format |
|---|---|---|
| TimeAxis_DCS | Time axis of the BFI data (Dbfit) | Column vector time axis of Dbfit (20Hz) |
| Dbfit | BFI value for each timepoint determined by fitting to the diffusion model | Column vector of data taken with source-detector separation 2.5cm |
| Marks_index | Marks in the data | Column 1: Original row number Column 2: Mark number Column 3: time relative to start of acquisition |
| raw_ABP_masterdata | ABP data from the CNAP | A four column vector, but only columns 1-4 are used. Column 1: Time from start of acquisition (125Hz) Column 2 Voltage data read directly off the ABP recording equipment by labview Column 3: ABP data, calculated in labview software. If cnap argument is called in pulse_analysis_main.m this column is updated from the raw voltage with the appropriate conversion upon loading. |
Data can be sorted into states and substates using an excel spreadsheet and the Marks_index.mat variable. Two different examples of state/substate naming are given as 'marks1' and 'marks2' spreadsheets and runfiles. The key terms are state_start, state_end, substate start, and substate end. Substate names must have the state name in them too. In the 'marks1' example, the state is a hypercapnia (HC) experiment and the substates are 'HC baseline' and 'HC Gas On'. The 'marks2' example is the same but uses a different naming scheme.
| HC State Start | SubjectID | Run | HC Baseline Substate Start | HC Baseline Substate End | HC Gas On Substate Start | HC Gas On Substate End | HC State End |
|---|---|---|---|---|---|---|---|
| 27_01_FC27 | 1 | 47 | 47 | 50 | 52 | 58 | 62 |
| 25_01_FC25 | 1 | 50 | 50 | 52 | 52 | 62 | 63 |
- In matlab, go to "Set Path" and select the "add folder with subfolders". Add the parent directory of this example to the search path.
- Open the desired run script. Three example scripts for running
pulse_analysis_main.mare given - one with no marks and two examples with marks. - Set filepaths such that
data_folderpoints to a folder with input data in the example format andsave_diris the desired output directory. If a marks spreadsheet is being used, set its path withanalysis table.
%%%%%%%%%%%%%%%%%%
% Set filepaths %
%%%%%%%%%%%%%%%%%%
BASEPATH = '/Users/taraurner/BuckleyLab/tmu_code/PIPELINES/20_HEALTHY_PULSATILITY_2023_BOE/PULSEANALYSIS/BOE_GithubExample'
data_folder = [BASEPATH filesep 'ExampleData']; % BFI and ABP data to be analyzed
analysis_tbl = [BASEPATH filesep 'data_index_usemarks2.xlsx'] % Contains marks input
save_dir = [BASEPATH filesep 'ExampleOutput' filesep 'marks2_pulse_analysis_out' filesep]; % Save location % Save location
if ~exist(save_dir,'dir')
mkdir(save_dir)
end- Set analysis options and run the main analysis function
pulse_analysis_main.
%%%%%%%%%%%%%%%%
% Run analysis %
%%%%%%%%%%%%%%%%
sds = 2;
fname = mfilename('fullpath');
ftext = fileread([fname '.m']);
text = [fname ':' ftext];
[ftbl,wtbl,ptbl,dat_tbl] = pulse_analysis_main(data_folder,save_dir,'sds',sds,...
'FileText',text,'WindowType','All','WindowSize',15,'AnalyzeMarks',analysis_tbl,'method',1,'corr_method','xcorr','cnap','true');Several tables are returned by the pulse_analysis_main function:
ftbl: table of full timeseries for all subjectswtbl: windows for all subjects, one row for each window in the dataptbl: all pulses identified for each subject, one row for each pulsedat_tbl: a mix of information from wtbl and ptbl with tables sorted by state and subject
Pulse morphology parameters in the table are as follows ('ln' stands for 'length normalized'):
params = {
'onset flow', 'osf_ln';...
'mean flow', 'mf_ln';...
'peak systolic flow', 'psf_ln';...
'end diastolic flow', 'edf_ln';...
'flow peak height', 'fph_ln';...
'flow area under the curve', 'f_auc_subedf';...
'flow resistive index', 'RI_ln';...
'flow pulsatility index', 'PI_ln';...
'onset pressure', 'osp_ln';...
'mean pressure', 'mp_ln';...
'peak systolic pressure', 'psp_ln';...
'end diastolic pressure', 'edp_ln';...
'pressure peak height', 'pph_ln';...
'pressure under the curve', 'p_auc_subedp';...
}
Pulse morphology quantification for each subject at rest and during hypercapnia can be found in PaperData/healthy-adult-pulsemorpho.csv.