Video Tutorial
Getting Started with
Pain2D-Tool
Pain2D-Tool allows to generate pain profiles (pain frequency maps), generating statistics for validation and comparison of the proposed models: Binary and KD-classifier.
Pain2D-Tool Tutorial
In this tutorial the use of Pain2D is explained step by step. The tutorial is suitable for everyone who is interested in analyzing pain drawings.
Before you start, download Pain2D and note the requirements of the program. For further information please see Download.
Setting template pain drawing
First start with uploading the template pain drawing as ZIP-file, which you created with Pain2D-Designer.
You can either select it with the “Browse..”-Button or just simply use Drag-and-Drop from your file system.
Then click on “Submit”.
You can now see the uploaded template on the web page.
Extraction of pain points from a pain drawing
The first function is the extraction of the pain points from the scanned pen-and-paper pain drawings. Select all of your pain drawings and upload them. Again u can use Drag-andDrop or the “Browse”-Button.
It is important to pseudonymize the scanned files. Please use the pattern “disease abbreviation_pain drawing number.pdf
! For more Details how to scan pain drawings visit the Collect data Tutorial.
Once the upload has finished, click on the “Submit”-Button.
You can see a progress bar in the bottom right corner of the window. This might take a while, so we just need to be patient.
If you got a warning, that some of the pain drawings are not readable, that means they are not properly filled in. For more Details see the ‘How to collect data tutorial’
If all your drawings are fine, you get a message says: “ Finished extraction. Please Download the files.”
You can now download the extracted points in a ZIP-file.
When you unpack the downloaded ZIP-file you can see it contains a corresponding JSON-file for each of the pain drawings.
Pain profiles
With the extracted pain points you can now create a pain profile.
To do that, select all of the JSON-files and upload them via the “Create pain profile”-area.
When the Upload is complete press the “Confirm”-Button.
Again, you can see the progress is running in the bottom right corner of the window.
You can now select between three different color-schemas: Gray, Blue-Yellow-Red or Red-Yellow-Blue.
Above the pain profile is a slider called “disease agreement”. This sets the percentage threshold for the amount of patients, that marked more then x% of the points. For example 31% will display all pain points that 31% or more patients have marked on pain drawings you just uploaded.
You can save your pain profile to not create it each time you start the program. For that click on the “Download data”-Button. The Pain profile will be saved as a csv-File.
By clicking on the “Download PNG”-Button you can also download the pain profile as a PNG-file.
After it finished downloading, you will see this image generated by Pain2D-Tool. This may be used and uploaded freely. So you can use this image wherever you want, for example in your publications.
CSV
PNG
You also have the option to upload an existing pain profile in a CSV-File.
Please remember, that you need to upload a template pain drawing first.
Simply select the pain profile (csv-File) in the “Load pain profile”-area and hit “confirm”.
The uploaded pain profile will be displayed and you can again choose a preferable color-schema.
AUC - Use of the binary classifier
If you need a binary prediction for two diseases you can use the AUC-tab in Pain2D-Tool.
Type in the name of both diseases you want to predict and upload the corresponding pain points as JSON files.
Pain2D-Tool is not a diagnostic tool. To get to know what the program is generating please read our publication (go to -> research).
Please keep in mind, that you need to extract the pain points first in the extraction tab. It is very important, that the name of the disease you gave is equal to the name of the JSON-files!
Click on the “Submit”-Button to generate the confusion matrix.
And here we see the fisher’s exact Test -results as well as the ROC with the AUC-value.
The test results can be downloaded as a csv-file when you click on the “Download CSV”-Button.
You can download the ROC – Image by just right-clicking on it and click on “save as”.
KDC - Use of the k-disease classifier
Similar to the AUC-function is the KDC-function, where you can make classifications of more then two diseases.
For that just click the “Add disease”-Button.
Again, the name of the disease you gave should be same as the name of the JSON-files!
Click on the “Submit”-Button. The files will be loaded and the result will be calculated, as you can see in the bottom right corner.
You can download the results by clicking the “Download CSV”-Button. The downloaded csv file should look like this.
The confusion matrix will be displayed as the result.
