Segmentation of the pulmonary lobes is relevant in clinical practice and particularly challenging for cases with severe diseases or incomplete fissures. In this work, an automated segmentation approach is presented that performs a transformation on Computed Tomography (CT) scans to subdivide the lungs into lobes. Content- based image retrieval has been a major research area with major focus on features extraction, due to its impact on image retrieval performance. When applying this in the medical field, required different feature extraction method that integrate some domain specific knowledge for effective image retrieval. Here a novel method called atlas based segmentation is proposed. Atlas methods usually require the use of image registration in order to align the atlas image or images to a new, unseen image. This method provides complementary information from past cases with confirmed diagnoses, to lung tissue classification and quantification in CT images. The system exploits the location of the pathological lung tissue and allows significant improvement in terms of early retrieval precision when compared to the approach based on global features only.