Related Experiment Video
Updated: May 19, 2026
09:31
Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone
Published on: April 8, 2015
Computerized quantification of bone tissue and marrow in stained microscopic images
Lin Shi1, Shangping Liu, Defeng Wang
1Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong.
Summary
This study introduces an automated algorithm for quantifying bone and marrow cell areas in histological images, crucial for diagnosing osteoporosis. The method accurately segments trabecular bone and marrow cell types, aiding physicians in disease assessment.
Area of Science:
- Histopathology
- Biomedical Image Analysis
- Osteoporosis Research
Background:
- Stained histological images are vital for diagnosing diseases like osteoporosis by revealing tissue architecture.
- Osteoporosis is linked to altered ratios of trabecular bone and bone marrow cells, with increased marrow fat content.
- Manual segmentation of histological images is time-consuming and labor-intensive.
Purpose of the Study:
- To develop an automatic algorithm for quantifying trabecular bone and marrow cell areas in histological images.
- To provide an efficient tool for analyzing bone tissue composition in osteoporosis research.
Main Methods:
- An automatic image segmentation algorithm was developed using mathematical morphological operations.
- The algorithm utilizes both color and morphological features of tissues.
- The method was implemented in Matlab and validated against manual segmentation.
Main Results:
- The proposed algorithm achieved high accuracy in quantifying trabecular bone and marrow cell areas.
- Validation showed a strong correlation (Pearson correlation coefficient > 0.94, P < 0.001) with expert manual segmentation.
- The algorithm effectively differentiates between trabecular bone, yellow marrow, and red marrow cells.
Conclusions:
- The developed automatic algorithm can effectively quantify stained bone histological images.
- This tool has the potential to significantly assist physicians in the diagnosis and assessment of osteoporosis.
- Automated segmentation offers a more efficient and objective approach compared to manual methods.
