Step 1. For our purpose, we obtained video files ".seq". To analyze our video files in MATLAB we need to convert ".seq" to "tiff".
Step 2. Respectively, inside the folder where ".seq" files are located, you should create a folder with the name "tiff". You should indicate the pathway for converted to "tiff" files to this folder.
Step 3. Now, next to the "tiff" folder create a new folder called "test".
Step 4. Take 6-10 tiff files and put it inside the folder "test". Usually, 6-10 files are enough to understand whether the chosen contrast value works or not for a particular mouse.
Step 5. Open MATLAB. We are going to utilize the first script that allows us to analyze movies with our data. In the main window open "browse for folder", then, open the folder that has a name "pupil segmentation" -> select folder. In "Current Folder" window click right mouse button -> select "add to path" -> current folder. Then select "eye_analysis_8_DM_CL8.m".
Step 6. In the new window where the script is located find a green "play" button on the control panel. It is called "run"
Step 7. Click it, then find your folder "test" and select this folder in order to test how applicable your contrast value to obtain the best version of graphs.
Step 8. When you select the folder "test" a window will appear. " please enter a value for contrast adjustment". You can enter different values. I used values from 0.03 to 0.15
Step 9. After MATLAB will be done with the analysis you see the graphs.
Step 9a. To save what is appeared in the "graph" window (it is better to save it cause it is not saved as a picture automatically by the program). So whenever you need a picture of a graph and you did not save it you have to analyze all tiffs for a particular set of videos again).
Step 9b. To save a picture from the window with graphs you can use File -> save as. Or make a window screenshot.
Step 10. If you satisfied with the appearance of the graph lines, its smoothness, and lack of artifacts, as well as the number of artifacts in general) you can apply the intensity of your choice to the main folder.
Step 10a. After step 10, repeat step 9, 9a, 9b.
After you analyzed all tiff files you needed in a folder you should have 3 files besides tiffs in a folder. "picture"
Step 11. Now we begin analysis of the average of each set of trials. After we will get the average we can compare averages. In my case, I had two groups of mice and compared average data between two groups.
Step 12. We need to start working with the script called "pupil_average_tg". Select the folder with this file and then select -> add to the path -> current folder
Step 13. Now we do a very important step - finding trials with big artifacts (if you have them) to make our average graph less noisy. To do this you need to click "open from a folder" sign on a "HOME" panel. Go to the folder where you saved 3 files (picture 1). Choose "diameter". Then you see on the left hand of the MATLAB window folders of this file. Choose folder diam keeper after "contrast_set"
Picture 2.
Step 14. You will see a table. Click on a column with number 1. Then choose "Plots" on the upper panel. You will see a separate graph for every trial. According to your parameters, you can choose to exclude a certain trail so you can get rid of artifacts and noise that make your data indeterminate.
Step 15. Choose what trials you are going to exclude according to your criteria. Write down the numbers of trials.
Step 16. Now we are ready to exclude bad trials and save the results. Choose "pupil_average_tg". Enter the bad trial numbers. Enter "1" if you want to save data. Your files for further analysis are ready!
Step 17. To analyze group pupil data use a file with the name "group_pupil_data". In our project, we had two intensities that we compared. It is indicated in the description of the script. You can have whatever intensities you need and follow the instruction in the script after you substitute your numbers.
Step 18. You can go farther and analyze data using a bar graph script. It is useful because you not only have a visual representation of the data but also a precise number of how one average data is different from another. Steps are described in the script.