Welcome to the InfoSci Platform
Reference Hub7
Lung Tumor Segmentation Using Marker-Controlled Watershed and Support Vector Machine

Lung Tumor Segmentation Using Marker-Controlled Watershed and Support Vector Machine

Surbhi Vijh, Rituparna Sarma, Sumit Kumar
Copyright: © 2021 |Volume: 12 |Issue: 2 |Pages: 14
ISSN: 1947-315X|EISSN: 1947-3168|EISBN13: 9781799861553|DOI: 10.4018/IJEHMC.2021030103
Cite Article Cite Article

MLA

Vijh, Surbhi, et al. "Lung Tumor Segmentation Using Marker-Controlled Watershed and Support Vector Machine." IJEHMC vol.12, no.2 2021: pp.51-64. http://doi.org/10.4018/IJEHMC.2021030103

APA

Vijh, S., Sarma, R., & Kumar, S. (2021). Lung Tumor Segmentation Using Marker-Controlled Watershed and Support Vector Machine. International Journal of E-Health and Medical Communications (IJEHMC), 12(2), 51-64. http://doi.org/10.4018/IJEHMC.2021030103

Chicago

Vijh, Surbhi, Rituparna Sarma, and Sumit Kumar. "Lung Tumor Segmentation Using Marker-Controlled Watershed and Support Vector Machine," International Journal of E-Health and Medical Communications (IJEHMC) 12, no.2: 51-64. http://doi.org/10.4018/IJEHMC.2021030103

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The medical imaging technique showed remarkable improvement in interventional treatment of computer-aided medical diagnosis system. Image processing techniques are broadly applied in detection and exploring the abnormalities issues in tumor detection. The early stage of lung tumor detection is extremely important in medical research field. The proposed work uses image processing segmentation technique for detection of lung tumor and the support vector classifier learning technique for predicting stage of tumor. After performing preprocessing and segmentation the features are extracted from region of lung nodule. The classification is performed on dataset acquired from national cancer institute for the evaluation of lung cancer diagnosis. The multi-class machine learning classification technique SVM (support vector machine) identifies the tumor stage of lung dataset. The proposed methodology provides classification of tumor stages and improves the decision-making process. The performance is evaluated by measuring the parameters namely accuracy, sensitivity, and specificity.