Related Experiment Video
Updated: Jul 8, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
Detection of spicules on mammogram based on skeleton analysis
1Graduate Sch. of Bio-Applications & Syst. Eng., Tokyo Univ. of Agric. & Technol.
IEEE Transactions on Medical Imaging
|January 1, 1996
Summary
This study introduces an image processing technique to detect spicules, key indicators of malignant tumors, on mammograms. The method effectively identifies radiating line patterns, improving diagnostic accuracy for breast cancer detection.
Area of Science:
- Medical Imaging
- Computer-Aided Diagnosis
- Oncology
Background:
- Spicules on mammograms are critical indicators of malignant tumors.
- Accurate detection of spicules is essential for early breast cancer diagnosis.
Purpose of the Study:
- To develop and evaluate a novel image processing method for detecting spicules on mammograms.
- To improve the accuracy of identifying tumor-related spicules.
Main Methods:
- Utilized line skeletonization to enhance spicule structures and reduce noise.
- Applied a modified Hough transform to identify radiating line patterns characteristic of spicules.
- Trained and tested the system on mammogram images.
Main Results:
- The proposed method successfully enhanced spicule axes and reduced other skeletal structures.
- Experiments demonstrated the effectiveness of the modified Hough transform in detecting radiating lines.
- Achieved a correct classification rate of 74% on a test set of 34 mammograms.
Conclusions:
- The developed image processing technique shows promise for accurate spicule detection in mammography.
- This method can aid in the early identification of malignant tumors by recognizing spicule patterns.
- Further validation may enhance its clinical utility in breast cancer diagnostics.
