Texture based feature extraction methods for content based medical image retrieval systems

Area of Science:

• Medical Imaging
• Computer Science
• Information Retrieval

Background:

• Content-based image retrieval (CBIR) systems are crucial for medical image archiving.
• Existing CBIR systems often lack efficiency for specialized medical image datasets.
• Spatial feature extraction methods are key to improving medical image retrieval.

Purpose of the Study:

• To evaluate the retrieval efficiency of spatial methods for medical image archiving.
• To compare the performance of Gray Level Co-occurrence Matrix (GLCM), Gray Level Run Length Matrix (GLRLM), and Gabor Wavelet for feature extraction.
• To identify the most effective method for medical image retrieval.

Main Methods:

• Utilized spatial feature extraction techniques including GLCM, GLRLM, and Gabor Wavelet.
• Constructed a database of hundreds of diverse medical images (brain, lung, sinus, bone).
• Experimentally assessed the retrieval efficiency rates of the investigated algorithms.

Main Results:

• Queries based on GLCM statistics yielded satisfactory results.
• Gabor Wavelet demonstrated superior effectiveness and accuracy compared to GLCM and GLRLM.
• The study confirmed the potential of spatial methods in enhancing medical image retrieval systems.

Conclusions:

• Gabor Wavelet is the most effective spatial method for content-based medical image retrieval.
• Further development of CBIR systems using advanced feature extraction is recommended for medical archiving.
• This research contributes to improving the efficiency and accuracy of medical image retrieval systems.