Is Your Pathology AI Really Ready?

New guidance explains why validating AI across scanners is essential for safe, reliable use in routine pathology

By
Jessica Allerton
August 18, 2026
8 min

The Pathologist

Overview

The Digital Pathology Association has published new recommendations addressing the validation of AI tools in pathology, emphasizing the need for separate validation of scanners and AI algorithms.

Background

As AI technology advances in pathology, its implementation has outpaced the establishment of necessary validation standards. Many healthcare organizations are deploying AI tools without confirming their compatibility with the scanners used for digital pathology images.

Data Highlights

No numerical or trial data provided in the source material.

Key Findings

AI tools in pathology must be validated for the specific scanners used to generate digital images.
Validation is crucial as AI models can lose accuracy when analyzing images from different scanners.
The recommendations provide a roadmap for laboratories to implement AI safely and consistently.
Scanner validation and AI validation should be conducted separately to identify performance issues accurately.
Quality control measures are essential for detecting technical failures in AI applications.