Segmentation of small structures in MR images: Semiautomated tissue hydration measurement

Abstract

Segmentation of small anatomic structures in noisy magnetic resonance (MR) images is inherently challenging because the edge information is contained in the same high-frequency image component as the noise. The authors overcame this obstacle in the analysis of the sural nerve in the ankle by processing images to reduce noise and extracting edges with an edge detection algorithm less sensitive to noise. Anatomic accuracy of the segmentation was confirmed by a neuroradiologist. A nerve hydration coefficient was determined from the signal intensity of the nerve in these segmented images. These semiautomated measurements of hydration agreed closely with those obtained with a previously described manual method (n = 44, P = .76). Each image in the study was analyzed identically, with no modification of the computer algorithm parameters. The data suggest that this robust method may be useful in a multicenter evaluation of diabetes treatment protocols.

Magnetic Resonance Imaging as a Biomarker in Diabetic and HIV-Associated Peripheral Neuropathy: A Systematic Review-Based Narrative

“…Question 2 asks whether the populations studied were clearly defined. Many of the earlier studies did not distinguish between type 1 and 2 diabetes, and often did not make clear from the methods section the number of patients from each group, or which patients were on insulin treatment (Koechner et al, 1995 ; Shibata et al, 1998 ; Wu et al, 2017 ; Edward et al, 2020 ) (14/18 complied). Question 8 asks whether different “levels” of exposure where compared, which in this context relates to whether groups were stratified based on severity or duration of neuropathy or diabetes, to make a more convincing link between disease and MR biomarkers (9/18 complied).…”

“…The earliest studies attempting to use MRI to non-invasively study peripheral nerves in DPN came from Eaton and Griffey in the late 1980s (Griffey et al, 1988 ; Koechner et al, 1995 ; Eaton et al, 1996 ). In a set of three papers, they used a proton-density-weighted Dixon sequence at 1.5 T as a measure of sural nerve water content.…”

“…They used water-only Dixon images, normalised to a copper-sulphate (CuSO4) control to derive an estimate of water content from proton density. They demonstrated higher nerve water content in patients with DPN compared with either diabetes without DPN, or healthy controls (Griffey et al, 1988 ; Koechner et al, 1995 ; Eaton et al, 1996 ). In their earlier study, they showed that sural nerve water content positively correlated with deficits on nerve conduction studies.…”

Virtual endoscopic environments in modern neurosurgical practice

“…A variety of approaches has been attempted to achieve automated segmentation [8] including techniques dependent on simple thresholding, [17] edge extraction, [24] clustering algorithms, [35] elastic matching, [47] fuzzy logic, [30] knowledge bases, [7] region growing, [18] and pattern recognition. [41] Clearly, the multiplicity of approaches devised is a byproduct of the intense research into the field of artificial intelligence.…”

Automated segmentation of the corpus callosum in midsagittal brain magnetic resonance images

“…[10][11][12][13][14] Related work includes a segmentation method using game theory, 15 segmentation using objective functions, 16 3-D segmentation of brain images, 17 detection of blood vessels in retinal images, 18 tumor detection, 19 segmentation and object recognition in echocardiograms, 20,21 and segmentation of cardiac MR images. 22 The corpus callosum, which is located at the center of the brain ͑Fig.…”