Recipe Description

The 3D EM Analysis in Aivia detects neuron structures in 3D electron microscopy images and creates surface reconstructions of the detection. The recipe applies a pre-trained deep learning model using the U-Net architecture to create a probability map for membrane and neuronal regions.

Parameters and Presets


Parameters and their descriptions are summarized in the table below.

Preset Group
Parameter Name
Min Value
Max Value
Intensity Threshold
255 (8-bit)
65,535 (16-bit)
Specifies the minimum intensity for segmentation on the deep learning results probability map; a lower value will detect larger objects as well as regions of higher uncertainty
Subset Filtering
Object Volume
1 x 1012
Specifies the range of objects to be included in the analysis results based on the volume of the detected objects
Subset Filtering
Smoothing Factor
Adjusts the amount of smoothing applied to the surface reconstructions of the detected objects; a lower value will generate surfaces with greater similarity to the input image


There are two preset groups in the recipe: Detection and Subset Filtering; each group has three pre-configured parameter grouping to help you get started on the analysis. The default preset values are as follows:


Parameter Name
Intensity Threshold
51 (8-bit)
13,107 (16-bit)
128 (8-bit)
32,768 (16-bit)
242 (8-bit)
62,258 (16-bit)

Subset Filtering

Parameter Name
Object Volume
5 - 40
20 - 500
10,000 - 5,000,000
Smoothing Factor


The 3D EM Analysis recipe generates morphological measurements for each detected 3D object. You can add additional measurements to the analysis results by using the Measurement Tool in Aivia. The measurements generated by the recipe are as follows.
  • Morphological
    • Surface Area
    • Volume
    • Convex Hull Volume
    • Volume Ratio


Before beginning the tutorial, please download the 3D EM Analysis Demo image ( For information on how to select presets or modify parameter values, please refer to the tutorial on how to use the Recipe Console. You can download the image directly here.

  1. Unzip the demo file and load the demo image, 3DEMAnalysisDemo.tif, into Aivia
  2. In the Recipe Console, click on the Recipe selection dropdown menu and select the 3D EM Analysis recipe
  3. Select the Low or Medium preset for the Detection preset group
  4. Click on the caret (>) to the left of the Subset Filtering preset group to show the preset parameters
  5. Change the Subset Filtering preset group parameter values to the following:
    • Object Volume: 100,000 - 80,000,000
    • Smoothing Factor: 2
  6. Click the From beginning button or press the F4 key on your keyboard to begin applying the recipe to the image

You will see two object set outputs: Cross Sections and Surfaces. The Cross Sections output allows you to view the cross section outline of the detected objects in 2D view; the Surfaces output allows you to view the surface reconstructions of the detected objects in 3D view.

2D Cross Sections (at z=5)

3D Surfaces

Image Credits

Narayanan "Bobby" Kasthuri and Jeff Lichtman, Harvard University,