An adaptive lung nodule detection algorithm

• Published in: 2009 Chinese Control and Decision Conference pp. 2361 - 2365
• Main Authors: Wei Guo, Ying Wei, Hanxun Zhou, DingYe Xue
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2009
• Subjects: an adaptive classification; Computed tomography; Detection algorithms; Feature extraction; Image databases; lung nodule detection; Lungs; modified Mahalanobis distance vector; Performance analysis; Radiology; Spatial databases; Support vector machine classification; Support vector machines; the Support Vector Machine
• ISBN: 9781424427222; 1424427223
• ISSN: 1948-9439; 1948-9447
• DOI: 10.1109/CCDC.2009.5192686

An adaptive lung nodule detection algorithm is presented in computed tomography (CT) images. Here, we present the details of the proposed algorithm and provide a performance analysis using a database from the department of radiology. Our algorithm consists of a feature selected part and a feature classified part. In the feature selected part, eight image features are extracted and Support Vector Machine (SVM) approach is applied to evaluate the classified performance of each feature. In the feature classified part, a nonlinear classifier is constructed on the basis of modified Mahalanobis distance. The adaptive algorithm is used to adjust the threshold in the classifier. The experiment indicated that the algorithm has a good sensitivity and accuracy for lung nodule detection.