# AI scoring of biomarkers and grade on pathology slides

Source: https://onco.cc/technologies/ai-pathology-scoring/  
OnCo record `ai-pathology-scoring` (Technology). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Software that reads digitised biopsy slides to detect cancer, grade it, and score biomarkers such as HER2, PD-L1 and Ki-67 more consistently than the eye alone.

## Summary

Deep-learning models trained on whole-slide images now detect prostate cancer in biopsies (Paige Prostate, the first FDA-authorised AI pathology tool, 2021), assist Gleason grading, count mitoses, and quantify immunohistochemistry such as HER2, PD-L1 and Ki-67, where pathologist agreement is known to be poor at the decision thresholds. Newer models infer molecular states, such as microsatellite instability, directly from routine H&E slides.

## Fields

- Kind: Technology
- Status: approved
- Last checked: 2026-09-04
- Principle: Convolutional or transformer networks trained on annotated whole-slide images produce per-region probabilities that are aggregated into detections, grades or scores presented to the pathologist.
- Since: 2021
- Strengths: Reduces scoring variability at treatment thresholds; Triages slides and flags missed foci; Can infer molecular features from H&E alone
- Limitations: Performance depends on scanner and stain domain; Regulatory clearance is product by product; Pathologist remains responsible for the diagnosis

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/Digital_pathology
- Wikipedia: https://en.wikipedia.org/wiki/Digital_pathology

## Connected records

- cancers: [HER2-positive breast cancer](https://onco.cc/cancers/breast-her2-positive/), [Non-small-cell lung cancer](https://onco.cc/cancers/nsclc/), [Prostate cancer](https://onco.cc/cancers/prostate/)
- fronts: [AI & Computation](https://onco.cc/fronts/ai-computation/), [Diagnostics & Biomarkers](https://onco.cc/fronts/diagnostics/)
- technologies: [Digital pathology & AI](https://onco.cc/technologies/digital-pathology-ai/), [Tumour mutational burden testing](https://onco.cc/technologies/tmb-testing/)
- companies: [Vicinity Bio](https://onco.cc/companies/vicinity-bio/)

---
JSON: https://onco.cc/api/v1/entities/ai-pathology-scoring.json