Related Experiment Video
Updated: May 18, 2026

Automated Slide Scanning and Segmentation in Fluorescently-labeled Tissues Using a Widefield High-content Analysis System
Published on: May 3, 2018
Evaluating Urine Cytology Slide Digitization Efficiency: A Comparative Study Using an Artificial Intelligence-Based
Jen-Fan Hang1,2, Yen-Chuan Ou3, Wei-Lei Yang4
1Department of Pathology and Laboratory Medicine, Taipei Veterans General Hospital, Taipei, Taiwan.
A novel heuristic scan method using AI significantly reduces scanning time and file size for digital cytology slides. This approach maintains high cell coverage rates, improving accuracy in diagnosing urothelial carcinoma.
Area of Science:
- Digital pathology
- Cytopathology
- Artificial intelligence in medicine
Background:
- Digitizing 3D cytology slides is challenging due to cell distribution and depth.
- Multi-Z-plane scanning is a solution but causes long scan times, large files, and complex review.
- Existing methods hinder clinical adoption of digital cytopathology.
Purpose of the Study:
- To introduce and evaluate a heuristic scan method as an AI-driven alternative to multi-Z-plane scanning for cytology slides.
- To compare the heuristic scan's efficiency and accuracy against conventional scanning techniques.
- To assess the impact of heuristic scanning on the interpretation of urothelial carcinoma cells.
Main Methods:
- Developed an AI-based heuristic scan simulation for cytology slide scanning.
- Compared heuristic scan with 21 Z-plane scans on 52 urine cytology slides across three cytopreparations.
- Analyzed cell coverage, scanning time, file size, and AI-aided interpretation accuracy.
Main Results:
- Heuristic scan achieved comparable SHGUC+ cell coverage rates to 5 Z-plane scans (0.78-0.91 vs. 0.75-0.88).
- Significantly reduced scanning time (137.2-635.0s vs. 332.6-1,278.8s) and file size (0.51-2.10GB vs. 1.16-3.10GB).
- Demonstrated higher accuracy in AI-aided interpretations compared to single Z-plane scans (62.5% vs. 37.5%).
Conclusions:
- Heuristic scan is a cost-effective alternative to multi-Z-plane scanning in digital cytopathology.
- It offers comparable cell capture efficiency with reduced time and file size.
- The method shows promise for improving the accuracy of digital urine cytology interpretations.
Related Concept Videos
Imaging Studies for Cardiovascular System V: CT
Imaging Studies for Cardiovascular System VI: Calcium -Scoring CT
Imaging Studies III: Computed Tomography
Imaging Studies VII: Vascular Imaging

