Computer-aided diagnostic (CAD) schemes have been developed to assist radiologists in the detection of various abnormalities in medical images such as lung nodules, intracranial aneurysms, and breast microcalcifications. In order to improve the sensitivity for detection of abnormalities, many researchers have employed filters for enhancement of a specific type of abnormality. However, these filters generally enhance not only the specific type of lesion, but also normal anatomic structures such as ribs, blood vessels, and airway walls. Therefore, lesions are often detected together with a large number of false positives caused by the normal anatomic structures. In this study, we developed selective enhancement filters for lung nodules, intracranial aneurysms, and breast microcalcifications, which can simultaneously enhance a specific type of lesion and suppress normal anatomic structures such as blood vessels and airway walls. Therefore, as preprocessing steps, these filters would be useful for improving the sensitivity of lesion detection and for reducing the number of false positives. We applied the selective enhancement filters to two-dimensional mammography, three-dimensional computed-tomography images, and magnetic resonance angiography (MRA) images to show its effectiveness for the enhancement of lung nodules, intracranial aneurysms, and breast microcalcifications.
All Science Journal Classification (ASJC) codes