Active Contour Based Segmentation and Classification for Pleura Diseases Based on Otsu’s Thresholding and Support...

Area of Science:

• Medical Imaging
• Computer-Aided Diagnostics
• Oncology

Background:

• Lung cancer diagnosis relies on early detection of pulmonary nodules.
• Accurate segmentation of lung nodules is crucial for computer-aided diagnostics.
• Distinguishing between normal and abnormal lung tissue presents a significant challenge.

Purpose of the Study:

• To propose an innovative method for identifying cancerous portions in lung images.
• To compare the efficacy of Otsu's segmentation and active contour segmentation for lung nodule analysis.
• To evaluate the performance of a Support Vector Machine (SVM) classifier in categorizing lung tissue.

Main Methods:

• Utilized Otsu's segmentation algorithm to initially identify potential cancerous regions.
• Employed active contour segmentation techniques for precise localization of lung nodules in CT images.
• Applied a Support Vector Machine (SVM) classifier to differentiate between normal and abnormal lung tissue post-segmentation.

Main Results:

• Both Otsu's thresholding and active contour segmentation were used to locate lung nodules in CT scans.
• The SVM classifier successfully categorized segmented portions as normal or abnormal.
• The proposed methods demonstrated suitability for achieving accurate segmentation and classification in complex lung images.

Conclusions:

• A comparative analysis was conducted between Otsu's segmentation and active contour methods coupled with an SVM classifier.
• The active contour segmentation combined with SVM classification yielded superior results for complex lung images compared to Otsu's method.
• This approach enhances the accuracy of lung nodule detection and classification in medical imaging.