Congress:
ECR24
Poster Number:
C-16469
Type:
EPOS Radiographer (scientific)
DOI:
10.26044/ecr2024/C-16469
Authorblock:
W. He, P. Xu, R. Xu, L. Huang, S-T. Feng, X. Li; Guangzhou , People's Republic of China/CN
Disclosures:
Weitao He:
Nothing to disclose
Ping Xu:
Nothing to disclose
Rulin Xu:
Nothing to disclose
Li Huang:
Nothing to disclose
Shi-Ting Feng:
Nothing to disclose
Xuehua Li:
Nothing to disclose
Keywords:
Artificial Intelligence, Gastrointestinal tract, Radioprotection / Radiation dose, CT-Enterography, Colonography CT, Image registration, Inflammation
Results: Compared to HIR, DLR with mild, standard and strong strengths resulted in significantly superior objective image quality (image noise, signal-to-noise ratio [SNR], contrast-to-noise ratio [CNR]), subjective image quality (image noise, contrast, sharpness), and diagnostic confidence for active bowel inflammation (all P<0.05). The sensitivity, specificity, and diagnostic accuracy for detecting active inflammation were 71.4% (75.0%, 70.0%), 83.3% (83.3%, 90.0%), and 75.0% (75.0%, 70.0%), respectively, for DLR with standard strength (DLR-mild and strong); while 43.3%, 80.0%, and 52.5.8% for HIR. The dichotomous inter-reader reliability (k) for the entire group was 0.81. The effective radiation dose of low dose CTE was 2.20 ± 0.29 mSv.