Abstract:White matter lesions (WMLs) are small groups of dead cells that clump together in the white matter of brain. In this paper, we propose a reliable method to automatically segment WMLs. Our method uses a novel filter to enhance the intensity of WMLs. Then a feature set containing enhanced intensity, anatomical and spatial information is used to train a random forest classifier for the initial segmentation of WMLs. Following that a reliable and robust edge potential function based Markov Random Field (MRF) is pro… Show more
“…Computer vision and pattern recognition have already been successfully applied to computer-aided diagnosis using MRI [17,18] and for segmentation of brain structures or lesions [19][20][21]. It has also been used to assess markers of SVD qualitatively.…”
Correspondence: Víctor González-Castro (victor.gonzalez@ed.ac.uk) or María del C. Valdés Hernández (M. Valdes-Hernan@ed.ac.uk) In the brain, enlarged perivascular spaces (PVS) relate to cerebral small vessel disease (SVD), poor cognition, inflammation and hypertension. We propose a fully automatic scheme that uses a support vector machine (SVM) to classify the burden of PVS in the basal ganglia (BG) region as low or high. We assess the performance of three different types of descriptors extracted from the BG region in T2-weighted MRI images: (i) statistics obtained from Wavelet transform's coefficients, (ii) local binary patterns and (iii) bag of visual words (BoW) based descriptors characterizing local keypoints obtained from a dense grid with the scale-invariant feature transform (SIFT) characteristics. When the latter were used, the SVM classifier achieved the best accuracy (81.16%). The output from the classifier using the BoW descriptors was compared with visual ratings done by an experienced neuroradiologist (Observer 1) and by a trained image analyst (Observer 2). The agreement and cross-correlation between the classifier and Observer 2 (κ = 0.67 (0.58-0.76)) were slightly higher than between the classifier and Observer 1 (κ = 0.62 (0.53-0.72)) and comparable between both the observers (κ = 0.68 (0.61-0.75)). Finally, three logistic regression models using clinical variables as independent variable and each of the PVS ratings as dependent variable were built to assess how clinically meaningful were the predictions of the classifier. The goodness-of-fit of the model for the classifier was good (area under the curve (AUC) values: 0.93 (model 1), 0.90 (model 2) and 0.92 (model 3)) and slightly better (i.e. AUC values: 0.02 units higher) than that of the model for Observer 2. These results suggest that, although it can be improved, an automatic classifier to assess PVS burden from brain MRI can provide clinically meaningful results close to those from a trained observer.
“…Computer vision and pattern recognition have already been successfully applied to computer-aided diagnosis using MRI [17,18] and for segmentation of brain structures or lesions [19][20][21]. It has also been used to assess markers of SVD qualitatively.…”
Correspondence: Víctor González-Castro (victor.gonzalez@ed.ac.uk) or María del C. Valdés Hernández (M. Valdes-Hernan@ed.ac.uk) In the brain, enlarged perivascular spaces (PVS) relate to cerebral small vessel disease (SVD), poor cognition, inflammation and hypertension. We propose a fully automatic scheme that uses a support vector machine (SVM) to classify the burden of PVS in the basal ganglia (BG) region as low or high. We assess the performance of three different types of descriptors extracted from the BG region in T2-weighted MRI images: (i) statistics obtained from Wavelet transform's coefficients, (ii) local binary patterns and (iii) bag of visual words (BoW) based descriptors characterizing local keypoints obtained from a dense grid with the scale-invariant feature transform (SIFT) characteristics. When the latter were used, the SVM classifier achieved the best accuracy (81.16%). The output from the classifier using the BoW descriptors was compared with visual ratings done by an experienced neuroradiologist (Observer 1) and by a trained image analyst (Observer 2). The agreement and cross-correlation between the classifier and Observer 2 (κ = 0.67 (0.58-0.76)) were slightly higher than between the classifier and Observer 1 (κ = 0.62 (0.53-0.72)) and comparable between both the observers (κ = 0.68 (0.61-0.75)). Finally, three logistic regression models using clinical variables as independent variable and each of the PVS ratings as dependent variable were built to assess how clinically meaningful were the predictions of the classifier. The goodness-of-fit of the model for the classifier was good (area under the curve (AUC) values: 0.93 (model 1), 0.90 (model 2) and 0.92 (model 3)) and slightly better (i.e. AUC values: 0.02 units higher) than that of the model for Observer 2. These results suggest that, although it can be improved, an automatic classifier to assess PVS burden from brain MRI can provide clinically meaningful results close to those from a trained observer.
“…Neuroimaging technology has made rapid advances over the past two decades revealing additional information about a number of neurologic processes. In particular, magnetic resonance imaging (MRI) has seen technological improvements in resolution, speed of acquisition, the addition of new sequences, and 3‐dimensional visualization of complex anatomy . Combined, these improvements have changed the way images are used in clinical settings and there are many efforts to establish more standardized evidence‐based uses that will inform clinical understanding of the extent, location, type, progression, and prognosis of an injury or disease process.…”
Section: Introductionmentioning
confidence: 99%
“…In particular, magnetic resonance imaging (MRI) has seen technological improvements in resolution, speed of acquisition, the addition of new sequences, and 3-dimensional visualization of complex anatomy. [1][2][3][4][5] Combined, these improvements have changed the way images are used in clinical settings and there are many efforts to establish more standardized evidence-based uses that will inform clinical understanding of the extent, location, type, progression, and prognosis of an injury or disease process. Current clinical use of medical imaging requires visual inspection and clinical judgment of competently trained radiologists.…”
While reliability between the two segmenting tools is fair to excellent, volumetric outcomes are statistically different between the two methods. As suggested by both developers, structure segmentation should be visually verified prior to clinical use and rigor should be used when interpreting results generated by either method.
“…The increase in the amount of input data without necessarily meaning an increase in the contextual or semantic data per se is known as data augmentation and has been used in brain image segmentation tasks. Several studies have introduced global spatial information as an additional input to CNN schemes in form of large 2D orthogonal patches downscaled by a factor(de Brebisson and Montana, 2015), integrated with intensity features from image voxels(Van Nguyen et al, 2015), as a number of hand-crafted spatial location features(Ghafoorian et al, 2016), synthetic volume(Steenwijk et al, 2013; Roy et al, 2015), or set of synthetic images that encode spatial information(Rachmadi et al, 2018b) for mentioning some examples. In other words, all input datasets are acquired under a limited set of conditions (e.g.…”
2 3 Magnetic resonance (MR) perfusion imaging non-invasively measures cerebral perfusion, 4 which describes the blood's passage through the brain's vascular network. Therefore it is widely 5 used to assess cerebral ischaemia. Convolutional Neural Networks (CNN) constitute the state-6 of-the-art method in automatic pattern recognition and hence, in segmentation tasks. But none 7 of the CNN architectures developed to date have achieved high accuracy when segmenting 8 ischaemic stroke lesions, being the main reasons their heterogeneity in location, shape, size, 9 image intensity and texture, especially in this imaging modality. We use a freely available CNN 10 framework, developed for MR imaging lesion segmentation, as core algorithm to evaluate the 11 impact of enhanced machine learning techniques, namely data augmentation, transfer learning 12 and post-processing, in the segmentation of stroke lesions using the ISLES 2017 dataset, which 13 contains expert annotated diffusion-weighted perfusion and diffusion brain MRI of 43 stroke 14 patients. Of all the techniques evaluated, data augmentation with binary closing achieved the 15 best results, improving the mean Dice score in 17% over the baseline model. Consistent with 16 previous works, better performance was obtained in the presence of large lesions. 17
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.