Background: Recently, cine magnetic resonance imaging (MRI) has shown promise for visualizing movement of the colonic wall, although assessment of data has been subjective and observer dependent. This study aimed to develop an objective and semi-automatic imaging metric of ascending colonic wall movement, using image registration techniques.

Methods: Cine balanced turbo field echo MRI images of ascending colonic motility were acquired over 2 min from 23 healthy volunteers (HVs) at baseline and following two different macrogol stimulus drinks (11 HVs drank 1 L and 12 HVs drank 2 L). Motility metrics derived from large scale geometric and small scale pixel movement parameters following image registration were developed using the post ingestion data and compared to observer grading of wall motion. Inter and intra-observer variability in the highest correlating metric was assessed using Bland-Altman analysis calculated from two separate observations on a subset of data.

Key Results: All the metrics tested showed significant correlation with the observer rating scores. Line analysis (LA) produced the highest correlation coefficient of 0.74 (95% CI: 0.55-0.86), p < 0.001 (Spearman Rho). Bland-Altman analysis of the inter- and intra-observer variability for the LA metric, showed almost zero bias and small limits of agreement between observations (-0.039 to 0.052 intra-observer and -0.051 to 0.054 inter-observer, range of measurement 0-0.353).

Conclusions & Inferences: The LA index of colonic motility derived from cine MRI registered data provides a quick, accurate and non-invasive method to detect wall motion within the ascending colon following a colonic stimulus in the form of a macrogol drink.

Download full-text PDF

Source
http://dx.doi.org/10.1111/nmo.12727DOI Listing

Publication Analysis

Top Keywords

colonic wall
8
ascending colonic
8
image registration
8
hvs drank
8
colon wall
4
wall motility
4
motility comparison
4
comparison novel
4
novel quantitative
4
quantitative semi-automatic
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!