Development of an improved CAD scheme for automated detection of lung nodules in digital chest images

TLDR

Lung cancer is the leading cause of cancer deaths in the U.S., with a 5‑year survival rate of only 13% that rises to 47% when detected early. This study developed an improved computer‑aided diagnosis scheme to automatically detect lung nodules in digital chest radiographs, aiming to reduce the 30% miss rate by radiologists. Using 200 PA chest radiographs (100 normal, 100 abnormal confirmed by radiologists), the CAD system first selects nodule candidates via multi‑threshold gray‑level difference imaging, classifies them into six groups, and then removes false positives with adaptive rule‑based tests and an artificial neural network. The CAD achieved 70% sensitivity with 1.7 false positives per image—substantially better than prior studies—and processes each image in about 20 s, indicating readiness for initial clinical evaluation.

Abstract

Lung cancer is the leading cause of cancer deaths in men and women in the United States, with a 5‐year survival rate of only about 13%. However, this survival rate can be improved to 47% if the disease is diagnosed and treated at an early stage. In this study, we developed an improved computer‐aided diagnosis (CAD) scheme for the automated detection of lung nodules in digital chest images to assist radiologists, who could miss up to 30% of the actually positive cases in their daily practice. Two hundred PA chest radiographs, 100 normals and 100 abnormals, were used as the database for our study. The presence of nodules in the 100 abnormal cases was confirmed by two experienced radiologists on the basis of CT scans or radiographic follow‐up. In our CAD scheme, nodule candidates were selected initially by multiple gray‐level thresholding of the difference image (which corresponds to the subtraction of a signal‐enhanced image and a signal‐suppressed image) and then classified into six groups. A large number of false positives were eliminated by adaptive rule‐based tests and an artificial neural network (ANN). The CAD scheme achieved, on average, a sensitivity of 70% with 1.7 false positives per chest image, a performance which was substantially better as compared with other studies. The CPU time for the processing of one chest image was about 20 seconds on an IBM RISC/6000 Powerstation 590. We believe that the CAD scheme with the current performance is ready for initial clinical evaluation.