This innovative technique could also track responses to cancer immunotherapies and refine clinical decision-making by identifying subcategories of diseases.

The research utilized an extensive dataset comprising over 16 million B cell and 25 million T cell receptor sequences from both healthy individuals and those with different immune states.

Findings revealed that T cell receptors are particularly informative for autoimmune diseases, while B cell receptors excel in identifying infections.

By combining results from T and B cells, the researchers significantly improved the accuracy of disease categorization, independent of demographic factors like age, sex, or race.

The newly developed machine learning-based test aims to enhance the diagnosis of autoimmune diseases by leveraging the immune system's memory of past diseases.

Current diagnostic methods often fail to utilize this immune memory effectively, as noted by lead author Maxim Zaslavsky.

The study, published on February 20, 2025, involved nearly 600 participants suffering from conditions such as COVID-19, lupus, and Type 1 diabetes.

Mal-ID has successfully identified disease states based on B and T cell receptor sequences, providing valuable insights for diagnosing complex autoimmune diseases.

Researchers at Stanford Medicine have developed a groundbreaking machine-learning technique known as Mal-ID, designed to diagnose various diseases by analyzing immune cell receptor sequences.

The researchers plan to adapt Mal-ID for a broader range of diseases, particularly focusing on the challenging area of autoimmune conditions.

The study received funding from several organizations, including the National Institutes of Health and the National Science Foundation, and involved contributions from various academic institutions.

The research team includes lead authors Zaslavsky and Erin Craig, with senior authors Scott Boyd and Anshul Kundaje.