Computer-aided Diagnosis of Pulmonary Infections Using Texture Analysis and Support Vector Machine Classification

Jianhua Yao, Andrew Dwyer, Ronald M. Summers, Daniel J. Mollura
2011 Academic Radiology  
Objective-The purpose of this study was to develop and test a computer-assisted detection method for identification and measurement of pulmonary abnormalities on chest CT in cases of infection, such as novel H1N1 influenza. The method developed could be a potentially useful tool for classifying and quantifying pulmonary infectious disease on CT. Subjects and Methods-Forty Chest CTs were studied using texture analysis and support vector machine (SVM) classification to differentiate normal from
more » ... normal lung regions on CT, including ten cases of immunohistochemistry proven infection, ten normal controls, and twenty cases of fibrosis. Results-Statistically significant differences in the receiver operator characteristics (ROC) curves for detecting abnormal regions in H1N1 infection were obtained between normal lung and regions of fibrosis, with significant differences in texture features of different infections. These differences enable quantification of abnormal lung volumes in CT imaging. Conclusion-Texture analysis and support vector machine classification can distinguish between areas of abnormality in acute infection and areas of chronic fibrosis, differentiate lesions having consolidative and ground glass appearances, and quantify those texture features to increase the precision of CT scoring as a potential tool for measuring disease progression and severity.
doi:10.1016/j.acra.2010.11.013 pmid:21295734 pmcid:PMC3061440 fatcat:obtovbenwzadlj7unsqhmdnof4