Lubomir M. Hadjiiski PhD, Presenter: Nothing to Disclose

David Zick, Abstract Co-Author: Nothing to Disclose

Heang-Ping Chan PhD, Abstract Co-Author: Institutional research collaboration, General Electric Company

Elaine M. Caoili MD, MS, Abstract Co-Author: Nothing to Disclose

Richard H. Cohan MD, Abstract Co-Author: Consultant, General Electric Company Consultant, Medscape, LLC

Chuan Zhou PhD, Abstract Co-Author: Nothing to Disclose

Kenny Heekon Cha MSc, Abstract Co-Author: Nothing to Disclose

To develop a CAD system for automated detection of ureter abnormalities in multi-detector row CT urography (CTU), which potentially can assist radiologists in detecting ureter cancer.

Our CAD system consists of two stages. In the first stage, tracking of the ureter is performed by previously developed COmbined Model-guided Path-finding Analysis and Segmentation System (COMPASS). After a user-input starting point, the ureter is automatically tracked by COMPASS based on anatomical knowledge and feature analysis of the contrast-filled lumen. In the second stage, the ureter wall is detected by using a polar transformation and tandem gray level thresholding to separate the ureter wall from the lumen and the background beyond the ureter wall. The ureter wall is considered abnormal if the wall thickness is greater than 1 mm. An upper limit of the wall thickness is imposed to reduce false positives. Finally, if the width of the thickened region extends to larger than 1/3 of the entire ureter wall circumference on a given slice it is considered a site of ureter wall thickening. In this pilot study, a limited data set of 38 patients (25 malignant and 13 benign) with biopsy-proven ureter wall thickenings was collected with IRB approval. Experienced radiologists identified 45 locations of ureter wall thickenings (28 cancers and 17 benign) on the CTU images as reference standard. The average lesion thickness was 4.0 mm (range: 1.8–9.2 mm). The average conspicuity rating was 3.1 (range: 2 to 5) on a scale of 1 to 5 (5 very subtle).

The COMPASS successfully tracked the ureters in all patients. 98% (44/45) of the ureter wall thickenings including 100% (28/28) of the ureter cancers were detected with 1.5 (55/38) false positives per patient. The missed benign wall thickening was a small lesion with thickness of 1.9 mm.

Our COMPASS and CAD system can track the ureter and detect ureter cancer of medium conspicuity and relatively small size. Further study is underway to collect a larger data set and improve the detection performance. This pilot study is a first step towards the development of a CAD system for detection of malignancy manifested as ureter wall thickening in CTU.

An accurate CAD system has the potential to assist radiologists in detection of ureter cancers at an early stage which usually are subtle in appearance.

Hadjiiski, L, Zick, D, Chan, H, Caoili, E, Cohan, R, Zhou, C, Cha, K, Computer Aided Detection of Ureter Wall Thickening in Multi-detector Row CT Urography.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045765.html