conference · 2025

CountPath: Automating Fragment Counting in Digital Pathology

Ana Beatriz Vieira, Maria João Valente, Tomé Albuquerque, Diana Montezuma, 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 counts tissue fragments on pathology slides more accurately and consistently than human experts.
Problem
Before analyzing pathology slides under a microscope, technicians must manually count the tissue fragments on each slide to ensure they match the original medical records. This quality control step is currently slow, labor-intensive, and prone to human error and inconsistency.
Method
The researchers developed CountPath, an automated system that uses advanced computer vision models, specifically YOLOv11 and Vision Transformers, to detect and count tissue fragments. These models are trained to recognize and count the fragments directly from digital images of the slides.
Results
The automated system achieved an accuracy of 90.1% in counting tissue fragments. This performance surpasses the typical accuracy range of human experts, which was found to be between 82% and 88%.
Takeaways
Automating fragment counting can significantly reduce the workload and subjectivity in pathology labs. The study shows that modern AI models can perform routine quality control tasks with higher consistency than human observers.
For industry
For pathology laboratories and medical technology providers, CountPath offers a reliable tool to automate a tedious manual quality control step. By integrating this software into digital pathology workflows, labs can speed up slide processing, reduce human error, and ensure higher diagnostic consistency.
Why it matters
This research contributes to safer and more efficient healthcare by minimizing the risk of diagnostic errors caused by missing tissue samples. Beyond pathology, the underlying technology could be adapted to automate quality control and object counting in other medical imaging and manufacturing fields.

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 YOLOv11 and Vision Transformer models. Our results demonstrate that the automated system achieves a level of performance comparable or even superior to that of experts, 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 90.1%, surpassing the range observed among experts (82–88%). This result further supports its suitability for integration into routine pathology workflows.

← All publications