preprint · arXiv (Cornell University) · 2025

CountPath: Automating Fragment Counting in Digital Pathology

Ana Beatriz Vieira, Maria João Valente, Diana Montezuma, Tomé Albuquerque, Liliana Ribeiro, Domingos Oliveira, João Monteiro, Sofia Gonçalves, Isabel Pinto, Jaime S. Cardoso, Arlindo L. Oliveira · 0 citations

View original publication

Summary AI-generated

TL;DR
Researchers have developed an automated system called CountPath that uses artificial intelligence to count tissue fragments on digital pathology slides as accurately as human experts.
Problem
Before analyzing medical slides under a microscope, pathologists must manually count tissue fragments to ensure they match the macroscopic report. This quality control step is crucial for accurate diagnosis, but manual counting is time-consuming, tedious, and highly subjective.
Method
The researchers developed an automated system that uses advanced computer vision models, specifically YOLOv9 and Vision Transformers. These models are trained to detect and count individual specimen fragments directly from digital pathology images.
Results
The automated system achieved an accuracy of 86% in counting tissue fragments. This performance falls directly within the 82% to 88% variation range observed among human experts, demonstrating that the AI is as reliable as manual assessment.
Takeaways
CountPath offers a reliable, efficient alternative to manual fragment counting in digital pathology. By matching human expert performance, it proves that deep learning can successfully automate routine quality control tasks in clinical workflows.
For industry
For digital pathology providers and clinical laboratories, CountPath offers a ready-to-integrate solution to automate a tedious pre-analytical quality control step. By replacing manual counting with AI, labs can reduce human error, standardize their workflows, and free up valuable time for pathologists.
Why it matters
This research advances patient safety by ensuring that no diagnostic tissue is overlooked or mislabeled during the biopsy preparation process. By streamlining the pre-analytical phase of digital pathology, this technology paves the way for faster, more reliable diagnoses and more efficient healthcare systems.

Abstract

Quality control of medical images is a critical component of digital pathology, ensuring that diagnostic images meet required standards. A pre-analytical task within this process is the verification of the number of specimen fragments, a process that ensures that the number of fragments on a slide matches the number documented in the macroscopic report. This step is important to ensure that the slides contain the appropriate diagnostic material from the grossing process, thereby guaranteeing the accuracy of subsequent microscopic examination and diagnosis. Traditionally, this assessment is performed manually, requiring significant time and effort while being subject to significant variability due to its subjective nature. To address these challenges, this study explores an automated approach to fragment counting using the YOLOv9 and Vision Transformer models. Our results demonstrate that the automated system achieves a level of performance comparable to expert assessments, offering a reliable and efficient alternative to manual counting. Additionally, we present findings on interobserver variability, showing that the automated approach achieves an accuracy of 86%, which falls within the range of variation observed among experts (82-88%), further supporting its potential for integration into routine pathology workflows.

← All publications