After After
Before Before

10163 - PD-L1, Germinal Center Detection, AI

DEVELOPED FOR OUTLINING GERMINAL CENTERS IN TONSILS

Tonsil tissue may be used as positive/negative controls for PD-L1 staining. Because the PD-L1 expression varies significantly between the tissue compartments, it is of interest to differentiate between tissue regions before evaluating the PD-L1 expression.

This APP automatically outlines germinal centers in tonsils stained for PD-L1 and is intended as an accessory APP to help isolate regions for further analysis.

KEYWORDS
Tonsil, Germinal Center, PD-L1, Quality Control, AI, Deep Learning

METHODS
The APP was developed using the using the DeepLabv3+ neural network available with Author™ AI. The neural network uses a cascade of layers of nonlinear processing units for feature extraction and transformation, with each successive layer using the output from the previous layer as input. DeepLabv3+ uses an encoder-decoder structure with atrous spatial pyramid pooling (ASPP) that is able to encode multi-scale contextual information by probing the incoming features with filters or pooling operations at multiple rates and multiple effective field-of-views. This means that instead of using step-wise upsampling blocks to incorporate features from different levels, this network only needs two upsampling steps, i.e. it is faster to train and analyze than e.g. the U-Net. All of this also means that the decoder module can refine the segmentation results along the object boundaries more precisely. For more information on the network architecture, see [1].

QUANTITATIVE OUTPUT VARIABLES
The output variables obtained from this protocol are:

  • Germinal Center Area [mm2]

AUXILIARY APPs
There are no auxiliary APPs available.

WORKFLOW
Step 1: Load and run the APP 01 Germinal Center Detection

STAINING PROTOCOL
There is no staining protocol available.

ADDITIONAL INFORMATION
To run the APP, a NVIDIA GPU with minimum 4 GB RAM is required.
The APP utilizes the Visiopharm Engine™, Engine™ AI and Viewer software modules, where Engine™ and Engine™ AI offer an execution platform to expand processing capability and speed of image analysis. The Viewer allows a fast review together with several types of image adjustment properties e.g. outlining of regions, annotations and direct measures of distance, curve length, radius, etc.
By adding the Author™ AI module the APP can be customized to fit other purposes. Author™ AI offers a comprehensive and dedicated set of tools for creating new fit-for-purpose analysis APPs, and no programming experience is required.

REFERENCES

LITERATURE
1. Chen, L., et. al., Encoder-decoder with atrous separable convolution for semantic image segmentation, Proceedings of the European conference on computer vision (ECCV) 2018, 801-818, DOI

RUO
FIGURE 1
FIGURE 1
Outlined germinal centers (blue) on TMA core.
FIGURE 2
FIGURE 2
Outlined germinal centers (blue) on TMA core.