Computerized detection of pulmonary nodules in digital chest images: Use of morphological filters in reducing false‐positive detections

TLDR

Radiologists miss up to 30 % of lung nodules in chest images. The study develops a computerized scheme that uses a difference‑image approach and morphological filtering to alert radiologists to suspected nodules and reduce missed detections. The scheme applies sequential erosion and dilation filters to extract features and mitigate rib and vessel interference during nodule detection. Adding the morphological feature‑extraction lowered the true‑positive rate by 13 % but cut false positives by 50 %, while improving overall sensitivity by roughly 50 % at three to four false positives per image.

Abstract

Currently, radiologists can fail to detect lung nodules in up to 30% of actually positive cases. If a computerized scheme could alert the radiologist to locations of suspected nodules, then potentially the number of missed nodules could be reduced. We are developing such a computerized scheme that involves a difference‐image approach and various feature‐extraction techniques. In this paper, we describe our use of digital morphological processing in the reduction of computer‐identified false‐positive detections. A feature‐extraction technique, which includes the sequential application of nonlinear filters of erosion and dilation, is employed to reduce the camouflaging effect of ribs and vessels on nodule detection. This additional feature‐extraction technique reduced the true‐positive rate of the computerized scheme by 13% and the false‐positive rate by 50%. In a comparison of the scheme with and without the additional feature‐extraction technique, inclusion of the additional technique increased the detection sensitivity by about half at the level of three to four false‐positive detections per chest image.