Value of Contrast-enhanced CT Scans in Diagnosis of Intestinal Obstruction Combined Intestinal Ischemia by Measuring Maximal Density of Region of Interest / 中国医学影像学杂志
Chinese Journal of Medical Imaging
; (12): 6-10, 2017.
Article
em Zh
| WPRIM
| ID: wpr-505656
Biblioteca responsável:
WPRO
ABSTRACT
Purpose To assess the feasibility of measuring maximal density of a region of interest (ROI) with contrast-enhanced CT in quantifying degrees of intestinal ischemia in patients with intestinal obstruction.Materials and Methods Abdominal CT images and reports of 160 patients with intestinal obstruction were retrospectively studied.All the data were reviewed by CT visual evaluation method and measuring maximal density of ROI respectively.The CT visual evaluation took the way of accumulated points,and divided the degrees of intestinal ischemia into five categories.The measuring maximal density of ROI quantified the degrees of bowel enhancement with a bar histogram on CT workstation.The results were compared with the pathological examination.The sensitivity,specificity,positive and negative predictive values and accuracy of the two methods were calculated respectively,and compared.Results The sensitivity,specificity,positive and negative predictive values and accuracy of CT visual evaluation method were 96.7%,72.9%,82.1%,94.4% and 86.2%,respectively.The sensitivity,specificity,positive and negative predictive values and accuracy of the measuring maximal density of ROI were 68.8%,100.0%,100.0%,71.4%,82.5%,respectively.By measuring the area under the ROC curve,the ROI method (0.995) was more accurate than CT visual evaluation method (0.908) in the diagnosis of bowel ischemia.Conclusion Measuring the maximal density of ROI can quantize bowel wall enhancement.It is a reliable and useful method in the diagnosis of intestinal ischemia,and in accordance with pathological results.
Texto completo:
1
Índice:
WPRIM
Tipo de estudo:
Diagnostic_studies
/
Prognostic_studies
Idioma:
Zh
Revista:
Chinese Journal of Medical Imaging
Ano de publicação:
2017
Tipo de documento:
Article