Aignostics Atlas Series: Atlas H&E-TME
Comprehensive analysis of the tumor microenvironment (TME)
Overview
Built using the foundation model we co-developed with Mayo Clinic, Atlas H&E-TME enables deep, accurate, and scalable spatial analyses of the TME in H&E-stained whole-slide images at single-cell resolution, empowering researchers with crucial insights for oncology drug development. Atlas H&E-TME has been extensively validated and provides qualitative polygon outputs that can be integrated into your preferred IMS, as well as thousands of higher order quantitative outputs per image.
Key advantages
Built with Atlas
Atlas H&E-TME is powered by Atlas, the industry-leading histopathology foundation model co-developed by Aignostics and Mayo Clinic.
Robust and rapid insights for every image
Atlas H&E-TME delivers comprehensive, validated analyses for cells, tissues, and higher order spatial networks, all within hours.
You control your data
Atlas H&E-TME is GDPR-compliant and follows ISO 27001 information security standards. Your data and results are protected during processing and transit, then automatically deleted from our network.
Features
Robust qualitative and quantitative outputs for a range of common cancer types.
Broad coverage for common cancers
Atlas H&E-TME currently supports analyses for bladder cancer, breast cancer, colorectal cancer, liver cancer, and lung cancer. Performance has been validated on over 90% of histologies by prevalence for each cancer type.
Robust slide quality control
QC models precisely flag areas that should be excluded from downstream analyses.
Cell detection and classification
Cell detection models flag individual cells and classify each into one of 9 different classes.
Tissue segmentation
Tissue segmentation models group relevant sections of each whole slide image into one of 7 different tissue regions.
Quantitative readouts
In addition to qualitative outputs, we provide over 5,000 quantitative metrics per image, including QC features and cell, tissue, and neighborhood-level details.
High-level metrics
endpoint reliability
Flexible access, tailored to your needs
We offer a variety of self-service and full-service options for accessing Atlas H&E TME:
- Licensing through our self-service API, with outputs viewable in your preferred IMS
- Licensing through our self-service API on Google Cloud Marketplace, with billing applied directly to your yearly cloud commitment
- Full-service licensing for customers who require additional support or prefer to use Aignostics’ in-house viewer
Atlas H&E-TME is for research use only. Not for use in diagnostic procedures.
Our cell classification approach works “out-of-the-box”
We compared how our cell classification model performed against a version that had been fine-tuned on additional samples. Adding additional training data did not improve cell classification model performance, indicating the model performs well "out-of-the-box."
