|
Lung Nodule Detection in CT
|
|||
|
Contemporary CT technology offers the potential of screening for the early detection of lung cancer. Proponents of existing screening trials argue that the highest chance of surgical cure from lung cancer lies in the detection of micronodular neoplasms (1-3 mm diameter lesions).
Current protocols using single slice spiral scanners and 5-10 mm thick slices are very successful in detecting larger nodules (> 3 mm diameter). However, reliable detection of micronodules requires overlapping, thin-section helical CT (hCT) scans performed with a multi-slice scanner. These image sequences generate extremely large volume data sets, consisting of 300-600 axial images, which are impractical to review in current radiology practice. Therefore, the advantages of sophisticated imaging technology cannot be fully utilized without the complementary development of efficient methods of image analysis and interpretation. Computer-assisted nodule detection is one such advance. Automated analytical programs that accurately distinguish focal lung pathology from normal bronchovascular anatomy can substantially improve the sensitivity of micronodule detection while reducing radiologist interpretation time. These programs are a necessary precondition for wide-scale, low-cost screening examinations for early lung cancer detection and surveillance. We propose a method to automatically identify lung nodules and micronodules from high-resolution hCT (HR-hCT) image data acquired from multi-slice scanners [Brow01b]. The method involves a model-based segmentation approach in which information about the size, shape, location, density and other properties of both normal and pathological structures will be used to automate the discrimination of focal lung nodules from normal bronchovascular anatomy. We are also developing an automated method for tracking nodules between serial exams [Brow01a]. | |||
|
|
|||
|
View examples of nodule detection: Return to Lung Cancer Page |
|||