Posts in AI Virtual and Digital Technologies
Prospective Risk Stratification for Lupus Nephritis in SLE: Retrospective Validation of a Multi-Layer Computational Engine

Lupus nephritis (LN) is a major cause of renal morbidity in SLE patients. To address the need for earlier and more individualized risk identification, we developed a computational engine incorporating longitudinal phenotype trajectories, treatment burden, and traditional biomarkers into a unified risk-scoring framework. The engine was retrospectively validated across a synthetic longitudinal cohort of 377 SLE patients, outputting a continuous composite score (0–100) and a four-tier risk classification: Low / Moderate / High / Critical. At the Moderate+ screening threshold, the engine achieved 100% sensitivity and 100% NPV within 90 days of confirmed LN diagnosis — meaning zero missed progressors. These findings suggest that this computational engine can identify patients approaching lupus nephritis up to 90 days before diagnosis, giving clinicians a critical window to intervene early, triage high-risk patients, and initiate nephrology referrals sooner.

Read More
Medical Record Natural Language Processing and Patient Biometric and Self-Reported Measures Inform an Artificial Intelligence Augmented Remote Care Management Strategy for Lupus Patients

This study combined natural language processing of medical records with patient biometric and self-reported data. Strong correlations identified 24 metrics predicting physician assessments, supporting AI-augmented remote lupus management strategies.

Read More
Expediting Systemic Lupus Erythematosus Classification of at-Risk Individuals Using Novel Technology: Outcomes of a Pilot Study

This study used a virtual/digital program to recruit at-risk individuals via online screening and sequential telehealth evaluations, classifying 18% with SLE in a mean of 371 days—shortening typical diagnosis time from 5-7 years—demonstrating potential for remote, accurate lupus assessment.

Read More
Tracking of Systemic Lupus Erythematosus (SLE) Longitudinally Using Biosensor and Patient-Reported Data: A Report on the Fully Decentralized Mobile Study to Measure and Predict Lupus Disease Activity Using Digital Signals-The OASIS Study

This decentralized study used machine learning to assess patient-reported outcomes, quality-of-life measures, and smartwatch biometric data from SLE patients. Models achieved significant predictive accuracy for disease flares, supporting proactive clinical screening.

Read More