HistoIndex hosted its signature Breakfast Meeting Event on November 8th, 2025 highlighting its multi-institutional collaboration with the National Institutes of Health (NIH) and Virginia Commonwealth University (VCU), where stain-free Second Harmonic Generation (SHG) and AI analytics are being applied to the NASH-CRN cohort to enable identification of fibrosis trajectories.

During this year’s AASLD  Breakfast Meeting, we were pleased to host an engaging session featuring two insightful presentations by Prof. Arun Sanyal and Dr. David Kleiner.

Prof. Sanyal opened the session with a talk on Predicting Fibrosis Progression in MASH Using Baseline SHG/AI-Derived Features.

Key Highlights:

• SHG/AI-DP from a single unstained biopsy can quantify fibrospatial and disease-activity related features.

• Preliminary data shows prognostic modeling of ≥1-stage fibrosis progression over multi-year follow-up using baseline SHG/AI-DP together with CRN stage, potentially serving as a risk-stratification tool.
  

• On the full cohort, adding qFibrosis features to baseline stage improves discrimination, demonstrating incremental value beyond conventional histology.

• Subsequent analysis will include interrogating discordant cases with differing fibrosis and NAS trajectories.

Dr. David Kleiner continued with HistoIndex Digital Pathology AI platform as an Assistive Tool for Aiding Pathologists.

Key points explored:

• SHG-based AI-aiding improved concordance among pathologists by up to 58% for fibrosis scoring.
  

• Reduced need for adjudication in clinical trials by ~25%, increasing efficiency.
  

• Improved reproducibility can increase study power and reduce sample size requirements (1400 → 800 subjects).
  

• AI helps highlight features not visible on traditional stains while maintaining expert oversight.

• Demonstrates strong value of AI-assisted scoring to support clinical trial decision-making and consistency.

Overall Takeaway

HistoIndex’s SHG/AI digital pathology platform demonstrates both prognostic and operational value:

• Quantitative fibrosis features could be used to predict disease progression.

• AI-assisted scoring enhances consistency and trial efficiency.

• These advancements support better patient selection, monitoring, and antifibrotic drug evaluation.