Use of AI-Enhanced Pattern Recognition and Autoantibody Discovery to Improve Anti-Nuclear Antibody Testing

Modernizing anti-nuclear antibody testing by combining AI-enhanced image analysis with multiplex autoantibody profiling to improve diagnostic precision and uncover novel autoimmune biomarkers.
Ongoing Other Autoantibodies Biological & Mechanistic Biomarker Discovery Data-Driven & Quantitative Disease Subtyping Experimental Platforms & Models Human Cohorts Immune Profiling Machine Learning & AI Translational & Clinical

Funding Year: 2025

Project Overview

Anti-nuclear antibody testing is an essential tool for diagnosing autoimmune diseases, but current methods are limited by subjective pattern recognition, coarse classification systems, and difficulty interpreting novel or mixed patterns. This project addresses those gaps by developing an AI-assisted tool to classify ANA patterns from expert-annotated clinical image datasets, and by identifying novel autoantibodies and antibody combinations that correspond to specific ANA patterns using custom multiplex assays. Leveraging a large archive of digital ANA images and discarded serum samples from the Clinical Immunology Lab, the study will build a powerful dataset linking visual ANA signatures to autoantibody specificity — enhancing diagnostic precision and enabling better clinical decision-making.

Impact & Innovation

AI-powered diagnostics for a foundational autoimmune test.

 

By pairing deep learning image classification with multiplex autoantibody profiling, this project transforms ANA testing from a subjective visual read into a precise, scalable diagnostic platform — with broad applicability across autoimmune disease settings.

  • Uncovers previously unrecognized ANA staining patterns and associated autoantibody signatures, redefining disease subtypes and revealing novel biomarkers in systemic autoimmunity
  • Generates high-value IP through a deep learning ANA classification model, nuclei segmentation tools, and customized multiplex panels positioned for commercialization as next-generation diagnostic tools
  • Advances the Consortium’s Integrated Data and Discovery Platforms pillar by building a scalable, AI-enabled clinical decision-support infrastructure for autoantibody testing across diverse settings
Research Approach

A framework designed for discovery

This project combines AI-driven digital image analysis with advanced multiplex autoantibody profiling to modernize ANA testing, linking visual staining patterns to autoantibody specificity and building a scalable diagnostic platform for autoimmune disease. The work integrates clinical data curation, machine learning model development, and translational tool validation.

Development of a deep learning-based ANA pattern classification model using expert-annotated clinical image datasets; nuclei segmentation and feature extraction pipelines applied to a large archive of digital ANA images; custom multiplex autoantibody assays applied to discarded serum samples; and integration of visual ANA signatures with autoantibody specificity data to build a linked diagnostic dataset.

Large archive of digital ANA images from the Clinical Immunology Lab, discarded serum samples with associated ANA patterns, expert-annotated clinical image datasets for model training, and multiplex autoantibody profiling datasets linking staining signatures to specific autoantibody combinations.

Development and validation of an AI-assisted ANA classification and autoantibody discovery platform, with translational goals including enhanced diagnostic accuracy, patient stratification, and clinical integration in collaboration with Immune Health and the Human Immunology Core. The platform is designed for commercialization as a next-generation diagnostic tool and clinical decision-support system across diverse autoimmune settings.

Investigators & Institutions

Powering the science

Principal Investigator

Eline T. Luning Prak, MD, PhD, Colton Consortium Member

Professor, Department of Pathology and Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania

Colton Centers & Institutions

Colton Center for Autoimmunity at the University of Pennsylvania

Perelman School of Medicine