Enhancement Filter for Computer-Aided Detection of Pulmonary Nodules on Thoracic CT images

Computer-aided detection (CAD) schemes can assist radiologists in the early detection of lung cancer which is crucial to the chance for curative treatment. Characterizing the pulmonary nodules in the multislice X-ray computed tomography (CT) images is notoriously difficult. This is due to the fact that the anatomical structures such as blood vessels, bronchi, and alveoli are subject to partial volume effects. Furthermore, the nodules connected with other dense anatomical structures increases the detection difficulties. In this paper, we propose a multiscale enhancement filter to improve the sensitivity for nodule detection, which is based on the undecimated wavelet transform and the eigenvalues of Yu matrix in multiplanar slices. As a preprocessing step of CAD for nodule detection, our enhancement filter can simultaneously enhance blob-like objects and suppress line-like structures. Therefore, it would be useful for reducing the number of false positives. We applied our enhancement filter to synthesized images and real medical images to demonstrate that it works well on enhancing a specific shape and suppressing other shapes. Our approach proposed in this paper is generic and can be applied for the analysis of blob-like structures in various other applications