Virtual Library
Start Your Search
Astrid Van Hoyweghen
Author of
-
+
P2.01 - Advanced NSCLC (Not CME Accredited Session) (ID 950)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 1
- Moderators:
- Coordinates: 9/25/2018, 16:45 - 18:00, Exhibit Hall
-
+
P2.01-92 - “All-In-One” Window: A New Way of Looking at Chest CT Studies in Thoracic Oncology (ID 12427)
16:45 - 18:00 | Author(s): Astrid Van Hoyweghen
- Abstract
Background
Historically, CT studies are always viewed in several window settings, optimized to evaluate specific anatomic structures and regions (mediastinal, lung, bone and vascular window). A newly developed image processing technique fuses these conventional windows into a single “all-in-one” window. This new window is specifically designed for comparison and follow-up of CT studies in oncology. The purpose of this study is to compare lesion detection on this “all-in-one” window versus conventional window settings.
a9ded1e5ce5d75814730bb4caaf49419 Method
In this retrospective study, 50 consecutive thoracic oncology chest CT examinations, containing 417 documented lesions and features, were reviewed by 6 radiologists, subdivided into 2 groups of 3 radiologists each, with similar levels of expertise in each group (experienced, junior and radiology resident). All scans were reviewed in conventional window settings (as in routine daily practice), by one group and in the “all-in-one” window by the other group. Lesions were listed as ‘missed’ when they were not seen by at least two out of three observers and as ‘well diagnosed’ when seen by at least two out of three observers.
4c3880bb027f159e801041b1021e88e8 Result
Out of the 417 lesions, 68 lesions were missed: 21 on the “all-in-one” window, 30 on conventional views and 17 on both views. Statistical analysis with linear mixed model showed that use of the “all-in-one” window did not result in an increase of missed lesions (p=1). Conversely, we found a tendency towards better lesion detection on the “all-in-one” window, though not strongly significant (p=0.07). Inter-observer agreement in both groups was similar (p=0.462).
8eea62084ca7e541d918e823422bd82e Conclusion
Our proposed new way of looking at chest CT images seems promising. In this study, we showed that lesion detection on a single "all-in-one" window is at least as good as on multiple conventional window settings. In further research, we will investigate the effect on lesion measurement and characterization.