Related Experiment Video
Updated: May 30, 2026
10:25
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
[Method of automatic detection of brain lesion based on wavelet feature vector]
Ya Fan1, Wei Liu, Huanqing Feng
1Department of Electronic Science and Technology, University of Science and Technology of China, Hefei 230027, China.
Summary
This study introduces an automated method for detecting brain lesions in CT images using wavelet feature vectors. The technique accurately distinguishes lesions from normal brain tissues, improving diagnostic capabilities.
Area of Science:
- Medical Imaging
- Computer-Aided Diagnosis
- Signal Processing
Background:
- Accurate detection of brain lesions is crucial for neurological diagnosis and treatment planning.
- Current methods for brain lesion detection can be labor-intensive and subjective.
- Automated techniques offer potential for improved efficiency and consistency.
Purpose of the Study:
- To develop and validate a novel automated method for brain lesion detection in CT images.
- To utilize wavelet feature vectors for enhanced lesion characterization.
- To improve the accuracy and efficiency of brain lesion identification.
Main Methods:
- Manual segmentation of normal CT images into gray matter, white matter, and cerebrospinal fluid.
- Fuzzy C-Means (FCM) clustering algorithm to obtain cluster centers for normal tissue types.
- Automatic segmentation of lesion-containing CT images into sub-images with controlled over-segmentation.
- Feature extraction using wavelet transforms and distance computation to classify sub-images as normal tissue or lesions.
Main Results:
- The proposed method demonstrated successful automatic detection of brain lesions.
- Experimental verification confirmed the efficacy of the wavelet feature vector approach.
- The method effectively differentiates between normal brain tissues and lesions.
Conclusions:
- The developed automated method shows promise for accurate and efficient brain lesion detection in CT imaging.
- Wavelet feature vectors provide a robust basis for distinguishing pathological tissue from normal brain structures.
- This approach has the potential to aid radiologists in clinical diagnosis.
