Medical Devices
Algorithms to Classify T Cell Activation by Autofluorescence Imaging
WARF: P190306US02
Inventors: Melissa Skala, Anthony Gitter, Zijie Wang, Alexandra Walsh
The Wisconsin Alumni Research Foundation (WARF) is seeking commercial partners interested in developing a method to non-invasively detect T cell activation by imaging NAD(P)H. These algorithms can be applied to NAD(P)H images taken with commercial imaging flow cytometers/sorters, and fluorescence microscopes.
Overview
The importance of T cells in immunotherapy is driving the development of technologies to better characterize T cells and improve therapeutic efficacy. One specific objective is assessing antigen-induced T cell activation because only functionally active T cells are capable of killing the desired targets. Autofluorescence imaging can assess functional activity of individual T cells in a non-destructive manner by detecting endogenous changes in metabolic co-enzymes such as NAD(P)H. However, recognizing robust patterns of T cell activity is computationally challenging in the absence of exogenous labels or information-rich autofluorescence lifetime measurements.
Non-invasive, contrast-agent free determination of T cell behavior is imperative for advancing studies of T cells in vivo and for quality control of clinical adoptive immune transfer procedures (e.g., CAR T cell therapies).
Non-invasive, contrast-agent free determination of T cell behavior is imperative for advancing studies of T cells in vivo and for quality control of clinical adoptive immune transfer procedures (e.g., CAR T cell therapies).
The Invention
Building on award-winning work, UW–Madison researchers have discovered that autofluorescence intensity images of NAD(P)H can accurately classify T cells as activated or not activated (‘naïve’ or ‘quiescent’), and have developed algorithms to classify T cell activation based on the images. Specifically, adapting pre-trained convolutional neural networks (CNNs) for the T cell activity classification task, T cells can be classified with 92 percent accuracy. These pre-trained CNNs perform better than classification based on summary statistics (e.g., cell size) or CNNs trained on the autofluorescence images alone.
This invention provides a way to non-invasively detect T cell activation by imaging NAD(P)H intensity. These algorithms can be applied to NAD(P)H images taken with commercial imaging flow cytometers / sorters, and fluorescence microscopes. If increased accuracy of T cell activation is needed for a specific application, additional measurements of the other NAD(P)H and FAD fluorescence endpoints can be obtained and used for classification.
This invention provides a way to non-invasively detect T cell activation by imaging NAD(P)H intensity. These algorithms can be applied to NAD(P)H images taken with commercial imaging flow cytometers / sorters, and fluorescence microscopes. If increased accuracy of T cell activation is needed for a specific application, additional measurements of the other NAD(P)H and FAD fluorescence endpoints can be obtained and used for classification.
Applications
- Screening and sorting activated T cells
Key Benefits
- Algorithms can be applied to existing commercial systems
- Method is non-invasive and contrast-agent free, hence can reduce reagent cost and variability
- Current methods to determine T cell activation require contrast agents and may require tissue/cell fixation
Stage of Development
The researchers have demonstrated that advanced machine learning models can accurately
classify T cell activity from NAD(P)H intensity images and that those image-based signatures transfer across human donors.
Additional Information
For More Information About the Inventors
Tech Fields
For current licensing status, please contact Jeanine Burmania at [javascript protected email address] or 608-960-9846