AI & Data Significance 6/10

Training and Validation Protocol for Primary Diagnosis With Digital Pathology: The Leeds Experience

Based on experience at Leeds Teaching Hospitals, which has scanned 100% of histology slides since 2018, this practical guide details the Royal College of Pathologists-endorsed training and validation protocol for digital primary diagnosis. The paper includes a comprehensive topographical list of diagnostic tasks and features that should form the basis of training slide sets, along with advice on common pitfalls for novice digital pathologists.

The original study

Practical guide to training and validation for primary diagnosis with digital pathology.

Authors
Williams BJ, Treanor D
Journal
Journal of clinical pathology
Type
Journal Article, Validation Study
PMID
31784420
Read the original study →

Original abstract

Numerous clinical pathology departments are deploying or planning to deploy digital pathology systems for all or part of their diagnostic output. Digital pathology is an evolving technology, and it is important that departments uphold or improve on current standards. Leeds Teaching Hospitals NHS Trust has been scanning 100% of histology slides since September 2018. In this practical paper, we will share our approach to training and validation, which has been incorporated into the Royal College of Pathologists' guidance for digital pathology implementation. We will offer an overview of the Royal College endorsed training and validation protocol and the evidence base on which it is based. We will provide practical advice on implementation of the protocol and highlight areas of digital reporting that can prove difficult for the novice digital pathologist. In addition, we will share a detailed topographical list of types of diagnostic tasks and features which should form the basis of digital slide training sets.