A scalable approach to T2-MRI colon segmentation

Orellana, Bernat; Monclús, Eva; Brunet, Pere; Navazo, Isabel; Bendezú, Álvaro; Azpiroz, Fernando

doi:10.1016/j.media.2020.101697

Cited by 8 publications

(13 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, studies of the volume of the adult colon have been carried out using magnetic resonance imaging (MRI). The advantages of MRI are excellent soft tissue contrast, cross sectional image quality, speed, and the lack of use of ionising radiation, which can provide comprehensive 3D views of each colonic segment (ascending, transverse, descending colon, and sigmoid-rectum) without preparation or use of contrast agents for the measurement of colonic volume [14][15][16]. Several studies in adults have demonstrated the responsiveness of the total colonic volume measured by MRI, which, in randomised placebo controlled trials, showed the expected increase in response to bulking agents like psyllium and kiwifruit [17,18].…”

Section: Introductionmentioning

confidence: 99%

Colonic Volume Changes in Paediatric Constipation Compared to Normal Values Measured Using MRI

et al. 2021

View full text Add to dashboard Cite

Background: Functional constipation in children is common. Management of this condition can be challenging and is often based on symptom reports. Increased, objective knowledge of colonic volume changes in constipation compared to health could provide additional information. However, very little data on paediatric colonic volume is available except from methods that are invasive or require unphysiological colonic preparations. Objectives: (1) To measure volumes of the undisturbed colon in children with functional constipation (FC) using magnetic resonance imaging (MRI) and provide initial normal range values for healthy controls, and (2) to investigate possible correlation of colonic volume with whole gut transit time (WGTT). Methods: Total and regional (ascending, transverse, descending, sigmoid, and rectum) colon volumes were measured from MRI images of 35 participants aged 7–18 years (16 with FC and 19 healthy controls), and corrected for body surface area. Linear regression was used to explore the relationship between total colon volume and WGTT. Results: Total colonic volume was significantly higher, with a median (interquartile range) of 309 mL (243–384 mL) for the FC group than for the healthy controls of 227 mL (180–263 mL). The largest increase between patients and controls was in the sigmoid colon–rectum region. In a linear regression model, there was a positive significant correlation between total colonic volume and WGTT (R = 0.56, p = 0.0005). Conclusions: This initial study shows increased volumes of the colon in children with FC, in a physiological state, without use of any bowel preparation. Increased knowledge of colonic morphology may improve understanding of FC in this age group and help to direct treatment.

show abstract

Section: Introductionmentioning

confidence: 99%

Colonic Volume Changes in Paediatric Constipation Compared to Normal Values Measured Using MRI

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Colonic biomass influences digestive sensations by two mechanisms: the volume of colonic content (both gas and faecal content) and the influence of microbiota on gut function. The effect of non-absorbable residues in the diet on gut sensations was investigated by a series of different studies using a method to measure gut content based on abdominal magnetic resonance imaging [56,57]. Using this methodology, it was shown that non-absorbable meal residues that enter the colon, increase the volume of colonic biomass [48].…”

Section: Non-absorbable Meal Residues Colonic Content and Microbiotamentioning

confidence: 99%

Gastrointestinal Contributions to the Postprandial Experience

Livovsky

Azpiroz

2021

Nutrients

Self Cite

View full text Add to dashboard Cite

Foodingestion induces homeostatic sensations (satiety, fullness) with a hedonic dimension (satisfaction, changes in mood) that characterize the postprandial experience. Both types of sensation are secondary to intraluminal stimuli produced by the food itself, as well as to the activity of the digestive tract. Postprandial sensations also depend on the nutrient composition of the meal and on colonic fermentation of non-absorbed residues. Gastrointestinal function and the sensitivity of the digestive tract, i.e., perception of gut stimuli, are determined by inherent individual factors, e.g., sex, and can be modulated by different conditioning mechanisms. This narrative review examines the factors that determine perception of digestive stimuli and the postprandial experience.

show abstract

“…Convolutional neural networks (CNN) and deep learning are recently conquering most fields of pattern recognition, so medical image segmentation cannot be an exception. CNNs were successfully applied in the segmentation of kidney abnormalities [54], prostate cancer [52], lesions caused by diabetic retinopathy [26,55] and melanoma [47], and the segmentation of liver [6], cardiac structures [8], colon [32], renal artery [50], mandible [57], bones [30], vertebrae [40], pancreas [59], lung [34], and hand vein network [22]. Also in the segmentation of brain tumors there are dozens of CNN based solutions.…”

Section: Introductionmentioning

confidence: 99%

A two-stage U-net approach to brain tumor segmentation from multi-spectral MRI records

Győrfi

Kovács

Szilágyi

2022

Acta Universitatis Sapientiae, Informatica

View full text Add to dashboard Cite

The automated segmentation of brain tissues and lesions represents a widely investigated research topic. The Brain Tumor Segmentation Challenges (BraTS) organized yearly since 2012 provided standard training and testing data and a unified evaluation framework to the research community, which provoked an intensification in this research field. This paper proposes a solution to the brain tumor segmentation problem, which is built upon the U-net architecture that is very popular in medical imaging. The proposed procedure involves two identical, cascaded U-net networks with 3D convolution. The first stage produces an initial segmentation of a brain volume, while the second stage applies a post-processing based on the labels provided by the first stage. In the first U-net based classification, each pixel is characterized by the four observed features (T1, T2, T1c, and FLAIR), while the second identical U-net works with four features extracted from the volumetric neighborhood of the pixels, representing the ratio of pixels with positive initial labeling within the neighborhood. Statistical accuracy indexes are employed to evaluate the initial and final segmentation of each MRI record. Tests based on BraTS 2019 training data set led to average Dice scores over 87%. The postprocessing step can increase the average Dice scores by 0.5%, it improves more those volumes whose initial segmentation was less successful.

show abstract

A scalable approach to T2-MRI colon segmentation

Cited by 8 publications

References 25 publications

Colonic Volume Changes in Paediatric Constipation Compared to Normal Values Measured Using MRI

Colonic Volume Changes in Paediatric Constipation Compared to Normal Values Measured Using MRI

Gastrointestinal Contributions to the Postprandial Experience

A two-stage U-net approach to brain tumor segmentation from multi-spectral MRI records

Contact Info

Product

Resources

About