Advanced AutoNeuron Settings
There are three sets of advanced settings that you can change for step 4 and step 5.
For most of your AutoNeuron tracing, the settings available from the workflow are sufficient.
In some cases, stacks may require fine-tuning some settings.
If you have questions about using the Advanced Settings, contact our Technical Support team.
Seed Detection
Sampling Density:AutoNeuron applies an invisible grid to sample the image uniformly.
- Coarse: less grid lines and less sampling performed.
- Medium: Optimizes density with speed, seed coverage, and final reconstruction accuracy.
- Dense: more grid lines and more sampling performed. Dense sampling typically leads to longer processing times to produce more seeds and increase the tracing coverage. But accuracy may not improve as a result.
Tracing
AutoNeuron determines the most suitable settings based on the characteristics of the image specified in Step 1.
Quick Set
Confocal High Magnification
Image type is fluorescent, maximum process width is more than 15 pixels. Super-ellipsoid detectors are enabled and usually improves the thickness estimates compared to just using tracing templates. Axial detection using tracing templates is enabled since axial signal in confocal images produce more reliable detections compared to in brightfield images. The movement of the tracing templates is also more relaxed due to more image pixels with signal compared to in low magnification images.
- Detect axial edges: On
- Use super ellipsoids: On, 20 iterations
- Rotation range: 5 x 11.25°
- Shifting: 5 pixels
Confocal Low Magnification
Image type is fluorescent, and maximum process width is less than 15 pixels. Super-ellipsoid detectors are disabled since low-magnification confocal images usually have insufficient amount of data for reliable detections. Axial detection using tracing templates is enabled since axial signal in confocal images produce more reliable detections compared to in brightfield images.
- Detect axial edges: On
- Use super ellipsoids: (same as Generic default: off)
- Rotation range: 5 x 11.25°
Brightfield
Image type is brightfield. Brightfield images usually have an out-of -focus axial signal, making axial detection using tracing templates unreliable. Super-ellipsoid detectors also rely on axial signals, and also have been found to be unreliable in brightfield images.
- Detect axial edges: Off
- Use super ellipsoids: Off
- Rotation Range: 5 x 11.25° (The Generic default is 3 x 11.25°).
Detector Movement Constraints
Used to set the Rotation and Shifting values.
Rotation
Generally, the tracing templates follow the process, rotating through the process. Any angle above the set amount is ignored. If you are growing neurons on a substrate, you may need to increase this value.
Shifting
While the tracing looks like a one-dimensional line, it really exists in three dimensions. Processes are not uniform in thickness. Shifting is the amount of leeway AutoNeuron uses as it moves from one point to the next. If you are working with irregular edges or thick branches, you may need to increase this value.
Branch Connections
You might have images or image stacks that are noisy, that is, that contain objects that may be mistaken for traces.
Instruct AutoNeuron to ignore traces based on segment length:
- Click the Ignore traces shorter than checkbox.
- Type a value OR click Measure to measure a trace in the image.
Branch Connection Criteria
Largest gap
Determines the largest gap AutoNeuron will "jump" to make a connection.
- Type a value or click Measure to measure it in the image.
Max deviation angle
Refers to the maximum angle AutoNeuron will consider when connecting branches. The value can be up to 180°.
- Type a value or click Measure to measure it in the image.
Min ratio of diameters
When two branches are considered for connection based on their respective location in (X, Y, Z) space, AutoNeuron compares the branch diameters of the end point of each branch. If the ratio of the diameters is greater than the specified value, AutoNeuron connects the two branches and recolors the tracing according to the branch that is closest to the soma (if the soma is traced).
- Use the slider or type a value to set the minimum ratio.
Example:
AutoNeuron compares Branch A and Branch B.
- Branch A's diameter at end point (DA) > Branch B's diameter at end point (DB).
- The default ratio is 50%.
AutoNeuron connects A and B if DB>= (DA/2)
Click Reset to reset all values. Click Load Defaults to load the AutoNeuron default settings.