scholarly journals Validation of Deep Learning-Based Artifact Correction on Synthetic FLAIR Images in a Different Scanning Environment

2020 ◽  
Vol 9 (2) ◽  
pp. 364 ◽  
Author(s):  
Kyeong Hwa Ryu ◽  
Hye Jin Baek ◽  
Sung-Min Gho ◽  
Kanghyun Ryu ◽  
Dong-Hyun Kim ◽  
...  
Keyword(s):  
Deep Learning ◽  
Periventricular White Matter ◽  
Artifact Correction ◽  
Significant Difference ◽  
Fluid Attenuated Inversion Recovery ◽  
Quantitative Analyses ◽  
Promising Solution ◽  
Test Sets ◽  
Magnetic Resonance Imaging Mri

We investigated the capability of a trained deep learning (DL) model with a convolutional neural network (CNN) in a different scanning environment in terms of ameliorating the quality of synthetic fluid-attenuated inversion recovery (FLAIR) images. The acquired data of 319 patients obtained from the retrospective review were used as test sets for the already trained DL model to correct the synthetic FLAIR images. Quantitative analyses were performed for native synthetic FLAIR and DL-FLAIR images against conventional FLAIR images. Two neuroradiologists assessed the quality and artifact degree of the native synthetic FLAIR and DL-FLAIR images. The quantitative parameters showed significant improvement on DL-FLAIR in all individual tissue segments and total intracranial tissues than on the native synthetic FLAIR (p < 0.0001). DL-FLAIR images showed improved image quality with fewer artifacts than the native synthetic FLAIR images (p < 0.0001). There was no significant difference in the preservation of the periventricular white matter hyperintensities and lesion conspicuity between the two FLAIR image sets (p = 0.217). The quality of synthetic FLAIR images was improved through artifact correction using the trained DL model on a different scan environment. DL-based correction can be a promising solution for ameliorating the quality of synthetic FLAIR images to broaden the clinical use of synthetic magnetic resonance imaging (MRI).

Brain magnetic resonance imaging does not contribute to the diagnosis of chronic neuroborreliosis

2007 ◽  
Vol 48 (7) ◽  
pp. 755-762 ◽  
Author(s):  
A. Aalto ◽  
J. Sjöwall ◽  
L. Davidsson ◽  
P. Forsberg ◽  
Ö. Smedby
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Disease Duration ◽  
Brain Magnetic Resonance Imaging ◽  
White Matter Lesions ◽  
Resonance Imaging ◽  
Mri Findings ◽  
Significant Difference ◽  
Fluid Attenuated Inversion Recovery ◽  
Magnetic Resonance Imaging Mri

Background: Borrelia infections, especially chronic neuroborreliosis (NB), may cause considerable diagnostic problems. This diagnosis is based on symptoms and findings in the cerebrospinal fluid but is not always conclusive. Purpose: To evaluate brain magnetic resonance imaging (MRI) in chronic NB, to compare the findings with healthy controls, and to correlate MRI findings with disease duration. Material and Methods: Sixteen well-characterized patients with chronic NB and 16 matched controls were examined in a 1.5T scanner with a standard head coil. T1- (with and without gadolinium), T2-, and diffusion-weighted imaging plus fluid-attenuated inversion recovery (FLAIR) imaging were used. Results: White matter lesions and lesions in the basal ganglia were seen in 12 patients and 10 controls (no significant difference). Subependymal lesions were detected in patients down to the age of 25 and in the controls down to the age of 43. The number of lesions was correlated to age both in patients (ρ = 0.83, P<0.01) and in controls (ρ = 0.61, P<0.05), but not to the duration of disease. Most lesions were detected with FLAIR, but many also with T2-weighted imaging. Conclusion: A number of MRI findings were detected in patients with chronic NB, although the findings were unspecific when compared with matched controls and did not correlate with disease duration. However, subependymal lesions may constitute a potential finding in chronic NB.

Automated Brain Imaging Diagnosis and Classification Model using Rat Swarm Optimization with Deep Learning based Capsule Network

2021 ◽  
pp. 2240001
Author(s):  
A. Vasantharaj ◽  
Pacha Shoba Rani ◽  
Sirajul Huque ◽  
K. S. Raghuram ◽  
R. Ganeshkumar ◽  
...  
Keyword(s):  
Deep Learning ◽  
Classification Model ◽  
Swarm Optimization ◽  
Diagnosis And Classification ◽  
Enhance Efficiency ◽  
Magnetic Resonance Imaging Mri ◽  
Learning Machine ◽  
The Brain

Earlier identification of brain tumor (BT) is essential to increase the survival rate of the patients. The commonly used imaging technique for BT diagnosis is magnetic resonance imaging (MRI). Automated BT classification model is required for assisting the radiologists to save time and enhance efficiency. The classification of BT is difficult owing to the non-uniform shapes of tumors and location of tumors in the brain. Therefore, deep learning (DL) models can be employed for the effective identification, prediction, and diagnosis of diseases. In this view, this paper presents an automated BT diagnosis using rat swarm optimization (RSO) with deep learning based capsule network (DLCN) model, named RSO-DLCN model. The presented RSO-DLCN model involves bilateral filtering (BF) based preprocessing to enhance the quality of the MRI. Besides, non-iterative grabcut based segmentation (NIGCS) technique is applied to detect the affected tumor regions. In addition, DLCN model based feature extractor with RSO algorithm based parameter optimization processes takes place. Finally, extreme learning machine with stacked autoencoder (ELM-SA) based classifier is employed for the effective classification of BT. For validating the BT diagnostic performance of the presented RSO-DLCN model, an extensive set of simulations were carried out and the results are inspected under diverse dimensions. The simulation outcome demonstrated the promising results of the RSO-DLCN model on BT diagnosis with the sensitivity of 98.4%, specificity of 99%, and accuracy of 98.7%.

White matter lesion burden in migraine with aura may be associated with reduced cerebral blood flow

Cephalalgia ◽  
2016 ◽  
Vol 37 (6) ◽  
pp. 517-524 ◽  
Author(s):  
Quan Zhang ◽  
Ritobrato Datta ◽  
John A Detre ◽  
Brett Cucchiara
Keyword(s):  
Blood Flow ◽  
Cerebral Blood Flow ◽  
White Matter ◽  
Magnetic Resonance ◽  
Migraine With Aura ◽  
Significant Difference ◽  
Fluid Attenuated Inversion Recovery ◽  
Weighted Magnetic Resonance Imaging ◽  
Magnetic Resonance Imaging Mri ◽  
And Control

Objective The objective of this study was to determine whether white matter hyperintensities (WMHs) in subjects with migraine are related to alterations in resting cerebral blood flow (CBF). Methods Migraine with aura (MWA), migraine without aura (MwoA), and control subjects were enrolled in a 1:1:1 ratio. WMH load was scored based on fluid-attenuated inversion recovery/T2-weighted magnetic resonance imaging (MRI) using a previously established semi-quantitative scale. Global and regional CBFs were quantified using arterial spin labelled perfusion MRI. Integrity of the circle of Willis was assessed with magnetic resonance angiography (MRA). Results A total of 170 subjects were enrolled (54 controls, 56 MWA, and 60 MwoA). There was no significant difference in subjects with ≥1 WMH across groups (22% controls, 29% MWA, 35% MwoA; p = NS). Similarly, high WMH load was not significantly different across groups (16.7% controls, 21.4% MWA, 25.0% MwoA; p = NS). High WMH load was strongly associated with increasing age (odds ratio: 1.08 per year, 95% confidence interval: 1.02–1.13, p = 0.01). Resting CBF was similar across groups, but was significantly higher in women. In MWA subjects with high WMH load, CBF was substantially lower ( p = 0.03). No association between WMH load and CBF was seen in control or MwoA subjects. Conclusions WHMs in MWA may be related to alterations in resting CBF.

scholarly journals Deep learning-based thin-section MRI reconstruction improves tumour detection and delineation in pre- and post-treatment pituitary adenoma

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Da Hyun Lee ◽  
Ji Eun Park ◽  
Yeo Kyung Nam ◽  
Joonsung Lee ◽  
Seonok Kim ◽  
...  
Keyword(s):  
Deep Learning ◽  
Pituitary Adenoma ◽  
Thin Section ◽  
Signal To Noise Ratio ◽  
Imaging Protocol ◽  
Normal Pituitary Gland ◽  
Significant Difference ◽  
Mri Reconstruction ◽  
Magnetic Resonance Imaging Mri ◽  
Higher Sensitivity

AbstractEven a tiny functioning pituitary adenoma could cause symptoms; hence, accurate diagnosis and treatment are crucial for management. However, it is difficult to diagnose a small pituitary adenoma using conventional MR sequence. Deep learning-based reconstruction (DLR) using magnetic resonance imaging (MRI) enables high-resolution thin-section imaging with noise reduction. In the present single-institution retrospective study of 201 patients, conducted between August 2019 and October 2020, we compared the performance of 1 mm DLR MRI with that of 3 mm routine MRI, using a combined imaging protocol to detect and delineate pituitary adenoma. Four readers assessed the adenomas in a pairwise fashion, and diagnostic performance and image preferences were compared between inexperienced and experienced readers. The signal-to-noise ratio (SNR) was quantitatively assessed. New detection of adenoma, achieved using 1 mm DLR MRI, was not visualised using 3 mm routine MRI (overall: 6.5% [13/201]). There was no significant difference depending on the experience of the readers in new detections. Readers preferred 1 mm DLR MRI over 3 mm routine MRI (overall superiority 56%) to delineate normal pituitary stalk and gland, with inexperienced readers more preferred 1 mm DLR MRI than experienced readers. The SNR of 1 mm DLR MRI was 1.25-fold higher than that of the 3 mm routine MRI. In conclusion, the 1 mm DLR MRI achieved higher sensitivity in the detection of pituitary adenoma and provided better delineation of normal pituitary gland than 3 mm routine MRI.

scholarly journals Lesion load assessment among multiple sclerosis patient using DIR, FLAIR, and T2WI sequences

2020 ◽  
Vol 51 (1) ◽  
Author(s):  
Abdullah Dhaifallah Almutairi ◽  
Hasyma Abu Hassan ◽  
Subapriya Suppiah ◽  
Othman I. Alomair ◽  
Abdulbaset Alshoaibi ◽  
...  
Keyword(s):  
Multiple Sclerosis ◽  
Neurological Disorders ◽  
Lesion Detection ◽  
Inversion Recovery ◽  
Daily Routine ◽  
Lesion Load ◽  
Significant Difference ◽  
Fluid Attenuated Inversion Recovery ◽  
Magnetic Resonance Imaging Mri ◽  
The Brain

Abstract Background Magnetic resonance imaging (MRI) is one of the diagnostic imaging modalities employing in lesion detection in neurological disorders such as multiple sclerosis (MS). Advances in MRI techniques such as double inversion recovery (DIR) made it more sensitive to distinguish lesions in the brain. To investigate the lesion load on different anatomical regions of the brain with MS using DIR, fluid attenuated inversion recovery (FLAIR) and T2-weighted imaging (T2WI) sequences. A total of 97 MS patients were included in our retrospective study, confirmed by neurologist. The patients were randomly selected from the major hospital in Saudi Arabia. All images were obtained using 3T Scanner (Siemens Skyra). The images from the DIR, FLAIR, and T2WI sequence were compared on axial planes with identical anatomic position and the number of lesions was assigned to their anatomical region. Results Comparing the lesion load measurement at various brain anatomical regions showed a significant difference among those three methods (p < 0.05). Conclusion DIR is a valuable MRI sequence for better delineation, greater contrast measurements and the increasing total number of MS lesions in MRI, compared with FLAIR, and T2WI and DIR revealed more intracortical lesions as well; therefore, in MS patients, it is recommended to add DIR sequence in daily routine imaging sequences.

scholarly journals Correlation of 4′-[methyl-11C]-thiothymidine PET with Gd-enhanced and FLAIR MRI in patients with newly diagnosed glioma

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Takashi Norikane ◽  
Katsuya Mitamura ◽  
Yuka Yamamoto ◽  
Yukito Maeda ◽  
Kenichi Tanaka ◽  
...  
Keyword(s):  
Standardized Uptake Value ◽  
Gadopentetate Dimeglumine ◽  
Tumor Proliferation ◽  
Newly Diagnosed ◽  
Enhanced Mri ◽  
Positron Emission ◽  
Pet Ct ◽  
Significant Difference ◽  
Fluid Attenuated Inversion Recovery ◽  
Magnetic Resonance Imaging Mri

Abstract Purpose To elucidate the biological association between tumor proliferation, tumor infiltration and neovascularization, we analyzed the association between volumetric information of 4′-[methyl-11C]thiothymidine (4DST) positron emission tomography (PET) and fluid-attenuated inversion recovery (FLAIR) and T1-weighted gadopentetate dimeglumine (Gd)-enhanced magnetic resonance imaging (MRI), in patients with newly diagnosed glioma. Methods A total of 23 patients with newly diagnosed glioma who underwent both 4DST PET/CT and Gd-enhanced MRI before therapy were available for a retrospective analysis of prospectively collected data. The maximum standardized uptake value (SUVmax) for tumor (T) and the mean SUV for normal contralateral hemisphere (N) were calculated, and the tumor-to-normal (T/N) ratio was determined. Proliferative tumor volume (PTV) from 4DST PET and the volume of Gd enhancement (GdV) and hyperintense region on FLAIR (FLAIRV) from MRI were calculated. Results All gliomas but 3 diffuse astrocytomas and one anaplastic astrocytoma had 4DST uptake and Gd enhancement on MRI. There was no significant difference between PTV and GdV although the exact edges of the tumor differed in each modality. The FLAIRV was significantly larger than PTV (P < 0.001). Significant correlations between PTV and GdV (ρ = 0.941, P < 0.001) and FLAIRV (ρ = 0.682, P < 0.001) were found. Conclusion These preliminary results indicate that tumor proliferation assessed by 4DST PET is closely associated with tumor-induced neovascularization determined by Gd-enhanced MRI in patients with newly diagnosed glioma.

scholarly journals Inter-Vendor Performance of Deep Learning in Segmenting Acute Ischemic Lesions on Diffusion-Weighted Imaging: A Multicenter Study

2020 ◽  
Author(s):  
Deniz Alis ◽  
Mert Yergin ◽  
Ceren Alis ◽  
Cagdas Topel ◽  
Ozan Asmakutlu ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Deep Learning ◽  
Diffusion Weighted Imaging ◽  
Ground Truth ◽  
Resonance Imaging ◽  
Validation Data ◽  
Clinical Applicability ◽  
Diffusion Weighted ◽  
Test Sets ◽  
Magnetic Resonance Imaging Mri

Abstract There is little evidence on the applicability of deep learning (DL) in segmentation of acute ischemic lesions on diffusion-weighted imaging (DWI) between magnetic resonance imaging (MRI) scanners of different manufacturers. We retrospectively included DWI data of patients with acute ischemic lesions from six centers. Dataset A (n=2986) and B (n=3951) included data from Siemens and GE MRI scanners, respectively. The datasets were split into the training (80%), validation (10%), and internal test (10%) sets and six neuroradiologist created ground-truth masks. Models A and B were the proposed neural networks trained on datasets A and B and also fine-tuned across the datasets using their validation data. Another radiologist performed the segmentation on the test sets for comparisons. The median Dice scores of models A and B were 0.858 and 0.857 for the internal tests, which were non-inferior to the radiologist’s performance, but demonstrated lower performance than the radiologist on the external tests. Fine-tuned models A and B achieved median Dice scores of 0.832 and 0.846, which were non-inferior to the radiologist's performance on the external tests. The present work shows that inter-vendor operability of deep learning for the segmentation of ischemic lesions on DWI might be enhanced via transfer learning; thereby, their clinical applicability and generalizability could be improved.

scholarly journals Retrospective motion artifact correction of structural MRI images using deep learning improves the quality of cortical surface reconstructions

NeuroImage ◽  
2021 ◽  
Vol 230 ◽  
pp. 117756
Author(s):  
Ben A Duffy ◽  
Lu Zhao ◽  
Farshid Sepehrband ◽  
Joyce Min ◽  
Danny JJ Wang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Structural Mri ◽  
Motion Artifact ◽  
Cortical Surface ◽  
Artifact Correction ◽  
Surface Reconstructions

scholarly journals Inter-vendor performance of deep learning in segmenting acute ischemic lesions on diffusion-weighted imaging: a multicenter study

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Deniz Alis ◽  
Mert Yergin ◽  
Ceren Alis ◽  
Cagdas Topel ◽  
Ozan Asmakutlu ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Deep Learning ◽  
Diffusion Weighted Imaging ◽  
Ground Truth ◽  
Resonance Imaging ◽  
Validation Data ◽  
Clinical Applicability ◽  
Diffusion Weighted ◽  
Test Sets ◽  
Magnetic Resonance Imaging Mri

AbstractThere is little evidence on the applicability of deep learning (DL) in the segmentation of acute ischemic lesions on diffusion-weighted imaging (DWI) between magnetic resonance imaging (MRI) scanners of different manufacturers. We retrospectively included DWI data of patients with acute ischemic lesions from six centers. Dataset A (n = 2986) and B (n = 3951) included data from Siemens and GE MRI scanners, respectively. The datasets were split into the training (80%), validation (10%), and internal test (10%) sets, and six neuroradiologists created ground-truth masks. Models A and B were the proposed neural networks trained on datasets A and B. The models subsequently fine-tuned across the datasets using their validation data. Another radiologist performed the segmentation on the test sets for comparisons. The median Dice scores of models A and B were 0.858 and 0.857 for the internal tests, which were non-inferior to the radiologist’s performance, but demonstrated lower performance than the radiologist on the external tests. Fine-tuned models A and B achieved median Dice scores of 0.832 and 0.846, which were non-inferior to the radiologist's performance on the external tests. The present work shows that the inter-vendor operability of deep learning for the segmentation of ischemic lesions on DWI might be enhanced via transfer learning; thereby, their clinical applicability and generalizability could be improved.

Sign in / Sign up

Export Citation Format

Share Document