Related Experiment Video
Updated: Jul 10, 2026
08:40
Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
Published on: April 8, 2016
Colour-texture based image analysis method for assessing the hormone receptors status in breast tissue sections
Spiros Kostopoulos1, Dionisis Cavouras, Antonis Daskalakis
1Medical Image Processing and Analysis Group, Laboratory of Medical Physics, School of Medicine, University of Patras, 26500 Rio, Greece. skostopoulos@upatras.gr
Summary
This study introduces a new image analysis method for assessing estrogen receptor (ER) status in breast cancer biopsies. The automated system achieved high accuracy, potentially improving objective ER status determination for hormonal therapy decisions.
Area of Science:
- Oncology
- Biomedical Engineering
- Computational Pathology
Background:
- Hormone receptor status is crucial for breast carcinoma prognosis and guiding hormonal therapy.
- Current determination relies on subjective visual assessment of stained tissue specimens.
- Objective and quantitative methods are needed to improve diagnostic consistency.
Purpose of the Study:
- To develop and assess a novel color-texture based image analysis methodology for quantitative evaluation of estrogen receptor (ER) positive status in breast carcinomas.
- To automate the assessment of ER status, reducing subjectivity in histopathological evaluation.
Main Methods:
- Utilized 22 immunohistochemically (IHC) stained breast biopsy specimens.
- Developed custom image analysis software employing color textural features.
- Implemented a k-Nearest Neighbor weighted votes classification algorithm for automated ER status assessment.
Main Results:
- The computer-based image analysis system achieved 86.4% overall accuracy.
- A Kendall's coefficient of concordance of 0.875 (p<0.001) indicated strong agreement.
- The system correctly ranked the ER positive status in 19 out of 22 cases.
Conclusions:
- Color-texture analysis of IHC stained specimens shows promise for the quantitative assessment of ER status in breast carcinomas.
- This automated approach may enhance objectivity and accuracy in predicting response to hormonal therapy.
- Further validation of this methodology could impact clinical decision-making in breast cancer management.
