Related Experiment Video
Updated: Jan 12, 2026
10:33
Expedited Radiation Biodosimetry by Automated Dicentric Chromosome Identification ADCI and Dose Estimation
Published on: September 4, 2017
16.5K
Automatic detection of sister chromatid exchanges using machine learning models and image analysis algorithms.
Mizuo Teraoka1, Shinya Matsumoto1, Ryotaro Kawasumi2
1Graduate School of Systems Design, Tokyo Metropolitan University, Asahigaoka 6-6, Hino-shi, 191 - 0065, Tokyo, Japan.
Scientific Reports
|November 6, 2025
Summary
This study introduces an automated system for analyzing sister chromatid exchange (SCE) using deep learning. The novel approach accurately measures SCEs, overcoming limitations of manual microscopic analysis.
Area of Science:
- Genetics
- Cell Biology
- Bioinformatics
Background:
- Sister chromatid exchange (SCE) is a recombination event between identical DNA sequences.
- SCE analysis is crucial in basic medicine and biology but is traditionally manual, time-consuming, and subjective.
- Existing methods for SCE analysis lack efficiency and objectivity.
Purpose of the Study:
- To develop a fully automated system for detecting and measuring sister chromatid exchanges (SCEs) from chromosome images.
- To improve the efficiency and objectivity of SCE analysis in biological and medical research.
Main Methods:
- Utilized Mask Region-based Convolutional Neural Network (Mask R-CNN) for accurate single chromosome detection and segmentation.
- Employed Vision Transformer (ViT) for classifying sister chromatid exchanges (SCEs).
- Integrated image processing and clustering algorithms for quantitative SCE measurement.
Main Results:
- The proposed system successfully automates the detection and measurement of SCEs.
- Achieved an accuracy of 84.10% in measuring the number of SCEs.
- Demonstrated a significant improvement over manual analysis in terms of speed and objectivity.
Conclusions:
- The developed deep learning-based system offers a robust and automated solution for SCE analysis.
- This technology has the potential to accelerate research in fields relying on SCE assessment, such as toxicology and cancer biology.
- Automated SCE analysis enhances reproducibility and reduces experimental bias.
