Detection of Pulmonary Nodules on CT and Volumetric Assessment of
Change over Time

Margrit Betke  
Computer Science Department, Boston College, Chestnut Hill MA 02167,
USA

Jane P. Ko
Department of Radiology, Massachusetts General Hospital, Harvard
Medical School, Boston, MA 02114, USA

We present a computer vision system that automatically detects
pulmonary nodules in computed tomography (CT) scans of oncology
patients, performs size We present a computer vision system that
automatically detects pulmonary nodules in computed tomography (CT)
scans of oncology patients, performs size analysis and assesses for
change in volume over time.  Thresholding, backtracking, and smoothing
algorithms have been developed to recognize the thorax and trace the
lung border. The regions within the lung that potentially contain
nodules are evaluated for their shape, size, and position. These
candidate regions are then characterized as nodules versus other
structures by comparing consecutive CT slices in the same study. A
preliminary system for the registration of studies has also been
developed. It estimates nodule volume in each study and evaluates the
volumetric change over time.  Our system has been tested on initial
and follow-up studies of four patients. Preliminary results were
detection of 284 nodules ranging between 1 and 32 mm at various
locations and assessment of their volumetric change over time.  Our
techniques have future applications for determining disease
progression, remission, and stability in oncologic patients in
addition to coregistration of different modalities within the thorax.