-
Notifications
You must be signed in to change notification settings - Fork 60
FAQs
Here are a few of the questions that are frequently asked about RRest. These are continually updated, so if you have a question of your own feel free to get in touch.
The algorithms should be relatively straightforward to run for the datasets provided. If you are having difficulty getting the algorithms to run then do have a look at the instructions provided for running the algorithms with the Synthetic Dataset here, which include an introductory video.
RRest can be used to run many RR algorithms (several hundred) on multiple input signals (such as ECG and PPG). However, you may only wish to run a subset of these algorithms, on a subset of input signals. To do so, modify the Universal Parameters accordingly.
For instance, you may only wish to run algorithms using a subset of the available feature measurement techniques. Perhaps you only want to use BW, AM and FM feature measurements. In this case you would modify up.al.options.FMe
, which can be found in the setup_universal_params.m script. Further instructions are given here.
You may also only wish to use a subset of the available input signals, such as just the ECG. In this case you would similarly modify up.al.options.extract_resp_sig
, within the setup_universal_params.m script. Further instructions are given here.
The synthetic dataset can be created using the Matlab ® script available here, which is within the overall script.
This will generated simulated ECG and PPG signals, subjected to the three respiratory modulations: baseline wander (BW), amplitude modulation (AM), and frequency modulation (FM).
If you only wish to generate a subset of these signals (such as only ECG, subject to only BW and AM), then you can do so by specifying the required signals and modulations in setup_params
function.
If you find that the processing is taking too long, then it may be helpful to try conducting a smaller set of processing to begin with. There are a couple of steps you can take to speed things up in the first instance:
-
Conduct the processing on a smaller number of subjects: You can discard subjects from a Matlab ® data file by editing the file in Matlab ®.
-
Use a subset of the available algorithms: The toolbox is set up to run a complete analysis over all the possible algorithms. However, some of the algorithms take much longer than others to run. Therefore, you can reduce the run-time considerably by following the instructions here. In particular, the pca technique specified in
up.al.options.FMe
takes a long time, so it would be prudent to eliminate this to begin with. Furthermore, you can eliminate the fusion step of the algorithms by removing the entry fromup.al.key_components
.
Part of the wider Respiratory Rate Estimation project
- Home
- Getting Started
- Input Data
- Universal Parameters
- Algorithm Structure
- Respiratory Signal Extraction
- Respiratory Rate Estimation
- Fusion of Respiratory Rates
- Signal Quality Assessment
- Reference Respiratory Rates
- Statistical Analysis
- Toolbox Versions
- System Requirements
- Known Issues
- Frequently Asked Questions