scholarly journals An Automatic Segmentation of T2-FLAIR Multiple Sclerosis Lesions

2008 ◽  
Author(s):  
Jean-christophe Souplet ◽  
Christine Lebrun ◽  
Nicholas Ayache ◽  
Gregoire Malandain
Keyword(s):  
Multiple Sclerosis ◽  
Test Data ◽  
Automatic Segmentation ◽  
Challenge Test ◽  
Flair Sequence ◽  
Conventional Mri ◽  
Hyperintense Signal ◽  
Automatic Methods ◽  
Mri Sequences

Multiple sclerosis diagnosis and patient follow-up can be helped by an evaluation of the lesion load in MRI sequences. A lot of automatic methods to segment these lesions are available in the literature. The MICCAI workshop Multiple Sclerosis (MS) lesion segmentation Challenge 08 allows to test and compare these algorithms. This paper presents a method designed to detect hyperintense signal area on T2-FLAIR sequence and its results on the Challenge test data. The proposed algorithm uses only three conventional MRI sequences: T1, T2 and T2-FLAIR. First, images are cropped, spatially unbiased and skull-stripped. A segmentation of the brain into its different compartments is performed on the T1 and the T2 sequences. From these segmentations, a threshold for the T2-FLAIR sequence is automatically computed. Then postprocessing operations select the most plausible lesions in the obtained hyperintense signals. Global result on the test data (80/100) is close to the inter-expert variability (90/100).

scholarly journals Using Convolutional Encoder Networks to Determine the Optimal Magnetic Resonance Image for the Automatic Segmentation of Multiple Sclerosis

2021 ◽  
Vol 11 (18) ◽  
pp. 8335
Author(s):  
Shaurnav Ghosh ◽  
Marc Huo ◽  
Mst Shamim Ara Shawkat ◽  
Serena McCalla
Keyword(s):  
Multiple Sclerosis ◽  
Deep Learning ◽  
Magnetic Resonance ◽  
Statistical Significance ◽  
Automatic Segmentation ◽  
Magnetic Resonance Images ◽  
Dice Similarity Coefficient ◽  
Learning Models ◽  
Mri Sequences ◽  
Convolutional Encoder

Multiple Sclerosis (MS) is a neuroinflammatory demyelinating disease that affects over 2,000,000 individuals worldwide. It is characterized by white matter lesions that are identified through the segmentation of magnetic resonance images (MRIs). Manual segmentation is very time-intensive because radiologists spend a great amount of time labeling T1-weighted, T2-weighted, and FLAIR MRIs. In response, deep learning models have been created to reduce segmentation time by automatically detecting lesions. These models often use individual MRI sequences as well as combinations, such as FLAIR2, which is the multiplication of FLAIR and T2 sequences. Unlike many other studies, this seeks to determine an optimal MRI sequence, thus reducing even more time by not having to obtain other MRI sequences. With this consideration in mind, four Convolutional Encoder Networks (CENs) with different network architectures (U-Net, U-Net++, Linknet, and Feature Pyramid Network) were used to ensure that the optimal MRI applies to a wide array of deep learning models. Each model had used a pretrained ResNeXt-50 encoder in order to conserve memory and to train faster. Training and testing had been performed using two public datasets with 30 and 15 patients. Fisher’s exact test was used to evaluate statistical significance, and the automatic segmentation times were compiled for the top two models. This work determined that FLAIR is the optimal sequence based on Dice Similarity Coefficient (DSC) and Intersection over Union (IoU). By using FLAIR, the U-Net++ with the ResNeXt-50 achieved a high DSC of 0.7159.

scholarly journals Walk Your Talk: Real-World Adherence to Guidelines on the Use of MRI in Multiple Sclerosis

Diagnostics ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 1310
Author(s):  
Mario Tortora ◽  
Mario Tranfa ◽  
Anna Chiara D’Elia ◽  
Giuseppe Pontillo ◽  
Maria Petracca ◽  
...  
Keyword(s):  
Multiple Sclerosis ◽  
Real World ◽  
University Hospital ◽  
Slice Thickness ◽  
Adequate Quality ◽  
Standardized Reporting ◽  
On Line ◽  
Mri Sequences ◽  
Weighted Sequences

(1) Although guidelines about the use of MRI sequences for Multiple Sclerosis (MS) diagnosis and follow-up are available, variability in acquisition protocols is not uncommon in everyday clinical practice. The aim of this study was to evaluate the real-world application of MS imaging guidelines in different settings to clarify the level of adherence to these guidelines. (2) Via an on-line anonymous survey, neuroradiologists (NR) were asked about MRI protocols and parameters routinely acquired when MS patients are evaluated in their center, both at diagnosis and follow-up. Furthermore, data about report content and personal opinions about emerging neuroimaging markers were also retrieved. (3) A total of 46 participants were included, mostly working in a hospital or university hospital (80.4%) and with more than 10 years of experience (47.9%). We found a relatively good adherence to the suggested MRI protocols regarding the use of T2-weighted sequences, although almost 10% of the participants routinely acquired 2D sequences with a slice thickness superior to 3 mm. On the other hand, a wider degree of heterogeneity was found regarding gadolinium administration, almost routinely performed at follow-up examination (87.0% of cases) in contrast with the current guidelines, as well as a low use of a standardized reporting system (17.4% of cases). (4) Although the MS community is getting closer to a standardization of MRI protocols, there is still a relatively wide heterogeneity among NR, with particular reference to contrast administration, which must be overcome to guarantee an adequate quality of patients’ care in MS.

scholarly journals Diffusion tensor imaging of the cervical spinal cord of patients with relapsing-remising multiple sclerosis: a study of 41 cases

2009 ◽  
Vol 67 (2b) ◽  
pp. 391-395 ◽  
Author(s):  
L. Celso Hygino Cruz Jr. ◽  
Romeu C. Domingues ◽  
Emerson L. Gasparetto
Keyword(s):  
Multiple Sclerosis ◽  
Spinal Cord ◽  
Fractional Anisotropy ◽  
Diffusion Tensor ◽  
Cervical Spinal Cord ◽  
Diffusion Tensor Mri ◽  
Mri Imaging ◽  
Conventional Mri ◽  
Mri Sequences ◽  
Diffusion Tensor Mr Imaging

OBJECTIVE: To evaluate the fractional anisotropy (FA) values of the multiple sclerosis (MS) plaques and normal-appearing cervical spinal cord (NASC) by diffusion tensor MRI imaging (DTI). METHOD: Forty-one patients with relapsing-remising MS and 37 controls were evaluated. All MRI exams were performed using a conventional protocol, as well as diffusion tensor MR imaging. Regions of interest were placed within the spinal cord lesions and in the normal appearing spinal cord adjacent to the plaque. RESULTS: The FA values were statistically reduced in the plaques compared to the surrounding NASC and to equivalent location in controls. A reduction in FA values was also observed in the spinal cord of MS patients without visible lesions on T2WI. CONCLUSION: We observed reduced fractional anisotropy in the demyelinating plaques and in the NASC of MS patients, corroborating the hypothesis that the histological extension of the MS lesions is more severe than the abnormalities seen in the conventional MRI sequences.

scholarly journals Diagnostic value of 3D-FLAIR MRI in children with sudden deafness caused by inner ear hemorrhage

2021 ◽  
Vol 4 (4) ◽  
pp. e000280
Author(s):  
Fu Sheng Gao ◽  
Xuan Jia ◽  
XiaoHui Ma ◽  
Jing Bi ◽  
Qiang Shu
Keyword(s):  
Inner Ear ◽  
Diagnostic Value ◽  
Term Treatment ◽  
Sudden Deafness ◽  
Short Term ◽  
Short Term Treatment ◽  
Flair Sequence ◽  
Mri Sequences ◽  
3D Flair

ObjectiveTo investigate the diagnostic value of three-dimensional fluid-attenuated inversion recovery (3D-FLAIR) MRI in children with sudden deafness caused by inner ear hemorrhage.MethodsThe diagnostic efficacies of three different MRI sequences in the examination of the inner ear for 32 children with sudden deafness were compared. Hearing examination results and 3-month follow-up outcomes were analyzed.ResultsThe age of 32 children with sudden deafness ranged from 5 to 18 years. MRI was performed from 1 to 18 days after onset. Six cases of sudden deafness caused by inner ear hemorrhage were finally diagnosed clinically. For different MRI sequences, the 3D-FLAIR sequence detected five positive cases; the conventional T1-weighted image sequence also detected five positive cases; but the conventional T2-weighted image sequence only detected three positive cases. The sensitivity and specificity of the 3D-FLAIR sequence in the diagnosis of inner ear hemorrhage were 83.3% (5/6) and 96.2% (25/26), respectively, and the area under the curve value of the receiver operating characteristic curve was 0.897. In the hemorrhage group, all six cases had extremely severe sensorineural hearing loss, and the hearing recovery was ineffective after 3 months of follow-up. The degree of hearing impairment, 3-month short-term treatment efficacy, and 3D-FLAIR MRI in the diagnosis of inner ear hemorrhage between hemorrhage group and non-hemorrhage group were statistically significant (p=0.043, p=0.000, p=0.000).Conclusions3D-FLAIR MRI is helpful for the diagnosis of inner ear hemorrhage in children with sudden deafness. Besides, short-term treatment indicates poor effects on children with severe hearing impairment.

Prediction of lymph node metastase in NSCLC

2010 ◽  
Vol 01 (05) ◽  
pp. 219-226 ◽  
Author(s):  
F. Beyer ◽  
B. Buerke ◽  
J. Gerss ◽  
K. Scheffe ◽  
M. Puesken ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Fdg Pet ◽  
Automatic Segmentation ◽  
Correlation Coefficients ◽  
Mediastinal Lymph Nodes ◽  
Intra Class Correlation ◽  
Axial Diameter ◽  
Pet Ct ◽  
Fdg Pet Ct

SummaryPurpose: To distinguish between benign and malignant mediastinal lymph nodes in patients with NSCLC by comparing 2D and semiautomated 3D measurements in FDG-PET-CT.Patients, material, methods: FDG-PET-CT was performed in 46 patients prior to therapy. 299 mediastinal lymph-nodes were evaluated independently by two radiologists, both manually and by semi-automatic segmentation software. Longest-axial-diameter (LAD), shortest-axial-diameter (SAD), maximal-3D-diameter, elongation and volume were obtained. FDG-PET-CT and clinical/FDG-PET-CT follow up examinations and/or histology served as the reference standard. Statistical analysis encompassed intra-class-correlation-coefficients and receiver-operator-characteristics-curves (ROC). Results: The standard of reference revealed involvement in 87 (29%) of 299 lymph nodes. Manually and semi-automatically measured 2D parameters (LAD and SAD) showed a good correlation with mean

Sign in / Sign up

Export Citation Format

Share Document