Knowledge Base

Data Analysis

Visualize Data

To open the Visualizations screen, choose Analysis > Visualizations, then click one of the tabs: Activity, Cardiology, or Flow.

Note: Only the subjects selected are visualized. See Analyze a Selection of Data.

Each row represents a measurement. If a subject is measured two or more times in the same video, data of that subject appears in a corresponding number of rows (see the column Measurement). The thumbnail shows a snapshot of the first video frame of the measurement with the area of interest highlighted in yellow.

Activity Plot

This is the time plot of activity (%) for Activity, Cardiology, and Flow. See How DanioScope Analyzes Video Images and Activity Settings.

Power Spectrum Plot

This is shown in the Cardiology tab. It describes how the activity signal is distributed over frequencies. The highest peak in the plot represents the dominant frequency, which is the heart rate in beats per minute.

Notes:

• A power spectrum is obtained by applying Discrete Fourier Transformation (DFT) to the samples in the activity signal.
• Important: Because the duration of the video is finite, there is a fixed limit to the frequency resolution (that is, the lowest frequency difference that can be detected), equal to 1/(video duration). For a 10-second video, this limit is 1/10 = 0.1 Hz. Therefore, when the dominant frequency in BPS reported for a 10-second video is 2.37 BPS, this actually means that the dominant frequency lies in the interval 2.32–2.42 BPS.
• If the dominant frequency is close to 0, it may result from artifacts in the video image (for example, flickering). Make sure that your video does not contain such artifacts.
• Tip: You can hide the Activity plot and/or the Power Spectrum plot by deselecting the check boxes at the top of the screen.
• Video is analyzed from the Start time to the Stop time defined in the Acquisition screen.

Play Back Video and Plots

Use playback to review a measurement in detail. Video is played back together with the activity plot.

1. Choose Analysis > Visualizations. Click one of the tabs: Activity, Cardiology, or Flow.
2. Click on the thumbnail for a measurement, or click one of the plots.
3. In the window that opens, click the buttons below the video or use the slider to play through or move through the video.

Data for that subject and measurement (highlighted in the video) are plotted together with the original video:

• For Activity: Activity Plot. The onset and offset activity thresholds for activity burst detection are also shown in the plot.
• For Cardiology: Activity Plot and Power Spectrum Plot.
• For Flow: Activity Plot.

Play the video frame-by-frame backward and forward with the corresponding buttons. Click the Loop button to play the video back repeatedly.

To Zoom In or Out a Plot

Drag around the area of the plot you want to zoom in. Always drag from left to right. To view the whole measurement, click and drag from right to left.

Notes:

• Tip: If you want to analyze activity peaks in detail, export the raw data.
• When you play the video, the hairline in the plot moves synchronously.
• The timeline of the plot represents the duration of the measurement, while the timeline of the video image represents the entire duration of the video. If your measurement is based on part of the video, when the video plays back you may see the hairline on the timeline of the video rather than on the plot. The same happens when you zoom in on a plot.
• To enlarge or reduce the video image, click the buttons at the upper-right corner of the video window.
• If you zoom in and play the video back, the hairline moves outside the selected time. Zoom out to view the rest of the timeline.

Export Plots

Do one of the following:

• To export the plot to a file, right-click a plot and select Save to File. Available format: Portable Network Graphics (*.png).
• To copy the plot to another program, right-click it and select Copy to Clipboard, then paste it into the other program.

Trial Statistics

Trial statistics summarize measurements for each subject. Choose Analysis > Trial Statistics. The Trial Statistics screen opens with results divided into pages for Activity, Cardiology, Flow, and Morphology. Each row in a table represents one subject. Trial statistics shows the mean and the standard error for each variable. N is the number of measurements per subject.

Note: Only the data selected is used in the analysis.

Activity

• Burst Activity (%). The percentage of the measurement duration the subject was scored as Active.
• Inactivity (%). Equals 100% minus Burst Activity (%).
• Total Burst Duration (s). The total duration of all instances when the subject was scored as Active, in seconds.
• Mean Burst Duration (s). The average duration of all instances when the subject was scored as Active, in seconds.
• Inactivity Duration (s). The total duration of the measurement minus the Total Burst Duration, in seconds.
• Burst Count. The number of times the subject was scored as Active.
• Burst Count / Minute. Equals Burst Count divided by the measurement duration, expressed as number per minute. Note that the measurement duration may not be the same as the length of the video, depending on the Start/Stop times of the measurement.

Cardiology

• BPS. The heart rate in number of beats per second.
• BPM. The heart rate in number of beats per minute.

Notes:

• Because the duration of the video is finite, there is a fixed limit to the frequency resolution (that is, the lowest frequency difference that can be detected), equal to 1/(video duration). For a 10-second video, this limit is 1/10 = 0.1 Hz. Therefore, when the dominant frequency reported for a 10-second video is 2.37 BPS, this actually means that the dominant frequency lies in the interval 2.32–2.42 BPS.
• If values of BPS are very low, they may result from artifacts in the video image (for example, flickering). Make sure that your video does not contain such artifacts.

Flow

• Flow Activity (%). The percentage of pixels within the area of interest which changed in gray scale value between subsequent frames in the video.

Morphology

The Morphology page shows the statistics for the measurements defined in the Experiment Settings.

• Body Length (mm)
• Eye Size (mm²)
• Pericardial Area (mm²)
• Somites. The number of somites observed in the image.
• Tail Detachment. With possible values: Positive and Negative.
• Body Curvature. With possible values: Dorsal, Ventral, and Straight.

Notes:

• <undefined> means that the measurement was not taken for that image.
• Only measurements taken on calibrated images are included in Trial Statistics.
• Make sure that all images have been associated with a calibration profile. In the Morphology screen, check the Calibration Profile column.
• If you measure Body Length with a polyline, Body Length is the total length of the segments that the polyline is made of.

Group Statistics and Charts

Group Statistics summarize measurements per group, where a group is defined based on the independent variables. Choose Analysis > Group Statistics and Charts.

Group Statistics gives you the average and standard error of the statistics calculated in Trial Statistics, averaged per group. Groups can only be defined based on the independent variables currently in the experiment. The charts visualize the group statistics with column bars. Scroll the horizontal bar to view all charts.

To Define Groups

1. Click the Group Settings button.
2. In the Group Settings window, select the independent variables you want to use to create groups, and how they are sorted. The values of the first variable appear on the x-axis of each chart. The values of the second variable appear in the series (indicated by the legend).
3. Click OK and then click the tab of your choice.

Notes:

• In the Group Statistics and Charts, N is the number of different subjects in that group, regardless of how many measurements are taken for each subject. If ten subjects are measured two times each, N is equal to 10.
• Important for Morphology: Make sure that all images taken with the same camera lens have the corresponding calibration profile. If not, group statistics may be incorrect. Only measurements taken on calibrated images are included in Group Statistics and Charts.
• To define additional independent variables, see Independent Variables.
• To assign the values of the new independent variable to subjects and measurements, do this in the measurement list of the corresponding screen (Activity, Cardiology, Flow, or Morphology).
• Click Swap IVs to exchange the independent variable between the x-axis and series.

To Compare Different Types of Measurements

To compare, for example, Burst Count per Minute with heart rate (BPS) in the same table:

1. Click the Comparison Settings button.
2. In the Comparison Settings window, select the measurements you want to compare.
3. Click OK and then click the Comparison tab for both the statistics and the charts.

To Export Group Charts

Do one of the following:

• To export the chart to a file, right-click a chart and select Save to File. Available format: Portable Network Graphics (*.png).
• To copy the chart to another program, right-click it and select Copy to Clipboard, then paste it into the other program.

Analyze a Selection of Data

Data selection is optional. You can use it to select the measurements associated with certain values of the independent variables. When visualizing data or calculating statistics, only the measurements specified in the data selection are used.

To select data, choose Analysis > Data Selection. In the Data Selection screen you can filter measurements for Activity, Cardiology, Flow, and Morphology separately. At the bottom of each of the four tables you can see how many of the measured subjects are currently selected.

To Select Measurements for Analysis

1. Select the check box under Filter for the independent variable of your choice. A window appears showing the independent variable values which you defined.
2. Select the values of your choice and click OK. The number of measured subjects in your selection is updated accordingly.
3. Repeat the steps above to refine your data selection based on other independent variables.
4. Choose one of the analysis options from the Analyze menu.

To Apply a Filter to All Measurement Types

Select the option "Apply changes in selected values of shared Independent Variables to all measurements" at the top of the Data Selection screen.

For instance, if you tested three groups of larvae (Control, Concentrate 1, and Concentrate 2) and you only select Control for Cardiology, this selection will also apply to the other measurements (for instance, Morphology).