Knowledge Base

Activity Settings

What Is Activity?

To detect embryo movements, DanioScope looks at the pixels within the shape around the embryo that change their gray scale value with time. At a given time, a pixel's gray scale value can range between 0 (pure black) and 255 (pure white).

Activity is defined as the proportion of pixels within the area of interest (embryo for Activity, heart area for Cardiology, and gut section for Flow) of video frame n, that change in gray scale value relative to the same area in the previous frame (n-1). Activity is expressed as the percentage of the changed pixels in the area of interest. A value of zero means that none of the pixels within the area of interest changed their gray scale value, like in a still image. A value of 100 means that all pixels in the area of interest have changed their value from frame n-1 to frame n.

Activity Onset and Offset

In any video file there is always a small proportion of pixels that change with time, even when the animal is not active. You must specify a minimum amount of pixel change, called Activity onset, above which the subject is scored as active. Furthermore, you must specify an Activity offset. When Activity (%) goes below the Activity offset value, the subject is scored as not active. This ends the time fragment scored as an activity burst (highlighted in red).

To refine detection of activity bursts, you have two extra options: the minimum time between peaks and the minimum peak duration.

To Adjust the Activity Settings

1. Select the video you want to analyze.
2. Click the Activity Settings button. The Activity Settings window opens.
3. The first subject/measurement S01/M1 is highlighted in the Subject/Measurement list, and in the video image.
4. Check that the peaks marked in red in the Activity plot correspond to true activity bursts. Use the playback controls to control the video. Zoom in the Activity plot if necessary. See Zoom In/Out.
   • To check for false negatives (the animal moves, but activity is not detected): position the video where activity bursts occur for that subject. Check in the Activity plot that the activity peak is highlighted in red. If that is not the case, reduce the Activity onset until the peak is highlighted.
   • To check for false positives (activity is detected, but the subject does not move): position the hairline in the Activity plot where an activity peak is highlighted in red. Check in the video that this corresponds to a true burst of activity. If that is not the case, increase the Activity onset until the peak is no longer highlighted.
5. Check the duration of single activity bursts. The Activity onset (green line) determines when the increase in activity is scored as the start of the activity burst (marked by the left side of the red rectangle). The Activity offset (orange line) determines when the decrease in activity is scored as the end of the activity burst (marked by the right side of the red rectangle). In most cases activity increases with a steep line and decreases more slowly. Therefore it makes sense to have Activity onset higher than Activity offset. If you want to capture most of the slow decrease in activity and include it in the burst duration, enter a lower Activity offset.

   Important: The left side of the red rectangle marks the first video frame when activity becomes higher than the onset. Similarly, the right side marks the first video frame when activity becomes lower than the offset. The start and stop times of the activity burst are based on the time stamps of those video frames. However, because the activity line shown in the plot is a linear interpolation between the actual (discrete) values of activity, a small difference may appear between the time that the line crosses the onset/offset and the true time of onset and offset (the sides of the red rectangle).

6. Check the time between single activity peaks. Check the distance between single peaks. If two adjacent peaks are counted as two bursts and you want them to be counted as one, increase the Min. Inter Peaks Interval. If two peaks are counted as one and you want them to be counted separately, reduce the Min. Inter Peaks Interval.
7. Set the minimum peak duration. Zoom in the activity plot (see Zoom In/Out) and check the duration of single peaks. If a short peak is detected as a separate burst and you want to exclude it from analysis, increase the Min Peak Duration.
8. Select another subject/measurement from the list and repeat steps 4 to 7. Try to find a value of Activity onset and offset that eliminates or minimizes the detection errors.
9. Click Apply to apply the new threshold to all measurements. After clicking Apply, the activity statistics are updated.
10. Check that the new thresholds also work well for other videos.

Notes

• Important: The new activity settings are used for all activity measurements in the experiment, including those already acquired.
• The Activity Settings button is only available after acquiring data from that video.

Activity Statistics

In the list below the video window you find the activity statistics per subject and measurement for that video file.

Note

The video frame corresponding to the Stop time is not analyzed. Video is analyzed from Start to Stop (included).

See Also

• Trial Statistics
• Group Statistics and Charts
• Export Statistics
• Export Raw Data

Source: DanioScope 1.1 Reference Manual - Activity, Noldus Information Technology