Automatic honeycomb lung segmentation in pediatric ct images

Since several lung diseases are diagnosed based on the patterns of lung tissue in medical images, texture segmentation is an essential part of the most computer aided diagnosis (CAD) systems. In this paper a novel composite method is proposed to segment the abnormality in lung tissue in pediatric CT images. The proposed approach is based on wavelet transform and intensity similarities. Our focus is on the honeycomb texture in lung tissue. After segmenting lung regions, wavelet transform is applied to decompose the image. The vertical subimage of lung is thresholded to extract high resolution areas. Then the regions with low pixel intensities are kept and grown to segment the honeycomb regions. The proposed method has been tested on 91 pediatric chest CT images containing healthy and unhealthy lung images. Statistical analysis has been done and the results show the sensitivity of 100% along with the specificity of 94.44%.