Publications
Results
- Showing results for:
- Reset all filters
Search results
-
Journal articleTarkin JM, Cole GD, Gopalan D, et al., 2020,
Multimodal imaging of granulomatosis with polyangiitis aortitis complicated by severe aortic regurgitation and complete heart block
, Circulation: Cardiovascular Imaging, Vol: 13, Pages: 1-3, ISSN: 1941-9651 -
Journal articleLi L, Wu F, Yang G, et al., 2020,
Atrial scar quantification via multi-scale CNN in the graph-cuts framework
, Medical Image Analysis, Vol: 60, ISSN: 1361-8415Late gadolinium enhancement magnetic resonance imaging (LGE MRI) appears to be a promising alternative for scarassessment in patients with atrial fibrillation (AF). Automating the quantification and analysis of atrial scars can bechallenging due to the low image quality. In this work, we propose a fully automated method based on the graph-cutsframework, where the potentials of the graph are learned on a surface mesh of the left atrium (LA) using a multi-scaleconvolutional neural network (MS-CNN). For validation, we have included fifty-eight images with manual delineations.MS-CNN, which can efficiently incorporate both the local and global texture information of the images, has been shownto evidently improve the segmentation accuracy of the proposed graph-cuts based method. The segmentation could befurther improved when the contribution between the t-link and n-link weights of the graph is balanced. The proposedmethod achieves a mean accuracy of 0.856 ± 0.033 and mean Dice score of 0.702 ± 0.071 for LA scar quantification.Compared to the conventional methods, which are based on the manual delineation of LA for initialization, our methodis fully automatic and has demonstrated significantly better Dice score and accuracy (p < 0.01). The method is promisingand can be potentially useful in diagnosis and prognosis of AF.
-
Journal articleWang Y, Yue W, Li X, et al., 2020,
Comparison Study of Radiomics and Deep Learning-Based Methods for Thyroid Nodules Classification Using Ultrasound Images
, IEEE Access, Vol: 8, Pages: 52010-52017Thyroid nodules have a high prevalence and a small percentage is malignant. Many non-invasive methods have been developed with the help of the Internet of Things to improve the detection rate of malignant nodules. These methods can be roughly categorized into two classes: radiomics based and deep learning based approaches. In general, convolutional neural networks based deep learning methods have achieved promising performance in many medical image analysis and classification applications; however, no existing comparison has been done between radiomics based and deep learning based approaches. Therefore, in this paper, we aim to compare the performance of radiomics and deep learning based methods for the classification of thyroid nodules from ultrasound images. On one hand, we developed a radiomics based method, which consists of extracting high throughput 302-dimensional statistical features from pre-processed images. Then dimension reduction was performed using mutual information and linear discriminant analysis respectively to achieve the final classification. On the other hand, a deep learning based method was also developed and tested by pre-training a VGG16 model with fine-tuning. Ultrasound images including 3120 images (1841 benign nodules and 1393 malignant nodules) from 1040 cases were retrospectively collected. The dataset was divided into 80% training and 20% testing data. The highest accuracies yielded on the testing data for radiomics and deep learning based methods were 66.81% and 74.69%, respectively. A comparison result demonstrated that the deep learning based method can achieve a better performance than using radiomics.
-
Journal articleLiu Y, Yang G, Hosseiny M, et al., 2020,
Exploring Uncertainty Measures in Bayesian Deep Attentive Neural Networks for Prostate Zonal Segmentation
, IEEE ACCESS, Vol: 8, Pages: 151817-151828, ISSN: 2169-3536- Author Web Link
- Cite
- Citations: 29
-
Journal articleAli A-RH, Li J, Yang G, 2020,
Automating the ABCD Rule for Melanoma Detection: A Survey
, IEEE ACCESS, Vol: 8, Pages: 83333-83346, ISSN: 2169-3536- Author Web Link
- Cite
- Citations: 10
-
Journal articleAli A-R, Li J, Kanwal S, et al., 2020,
A Novel Fuzzy Multilayer Perceptron (F-MLP) for the Detection of Irregularity in Skin Lesion Border Using Dermoscopic Images.
, Front Med (Lausanne), Vol: 7, ISSN: 2296-858XSkin lesion border irregularity, which represents the B feature in the ABCD rule, is considered one of the most significant factors in melanoma diagnosis. Since signs that clinicians rely on in melanoma diagnosis involve subjective judgment including visual signs such as border irregularity, this deems it necessary to develop an objective approach to finding border irregularity. Increased research in neural networks has been carried out in recent years mainly driven by the advances of deep learning. Artificial neural networks (ANNs) or multilayer perceptrons have been shown to perform well in supervised learning tasks. However, such networks usually don't incorporate information pertaining the ambiguity of the inputs when training the network, which in turn could affect how the weights are being updated in the learning process and eventually degrading the performance of the network when applied on test data. In this paper, we propose a fuzzy multilayer perceptron (F-MLP) that takes the ambiguity of the inputs into consideration and subsequently reduces the effects of ambiguous inputs on the learning process. A new optimization function, the fuzzy gradient descent, has been proposed to reflect those changes. Moreover, a type-II fuzzy sigmoid activation function has also been proposed which enables finding the range of performance the fuzzy neural network is able to attain. The fuzzy neural network was used to predict the skin lesion border irregularity, where the lesion was firstly segmented from the skin, the lesion border extracted, border irregularity measured using a proposed measure vector, and using the extracted border irregularity measures to train the neural network. The proposed approach outperformed most of the state-of-the-art classification methods in general and its standard neural network counterpart in particular. However, the proposed fuzzy neural network was more time-consuming when training the network.
-
Journal articleZhuang X, Li L, Payer C, et al., 2019,
Evaluation of algorithms for multi-modality whole heart segmentation: An open-access grand challenge
, Medical Image Analysis, Vol: 58, ISSN: 1361-8415Knowledge of whole heart anatomy is a prerequisite for many clinical applications. Whole heart segmentation (WHS),which delineates substructures of the heart, can be very valuable for modeling and analysis of the anatomy and functionsof the heart. However, automating this segmentation can be challenging due to the large variation of the heart shape,and different image qualities of the clinical data. To achieve this goal, an initial set of training data is generally neededfor constructing priors or for training. Furthermore, it is difficult to perform comparisons between different methods,largely due to differences in the datasets and evaluation metrics used. This manuscript presents the methodologiesand evaluation results for the WHS algorithms selected from the submissions to the Multi-Modality Whole Heart Segmentation (MM-WHS) challenge, in conjunction with MICCAI 2017. The challenge provided 120 three-dimensionalcardiac images covering the whole heart, including 60 CT and 60 MRI volumes, all acquired in clinical environmentswith manual delineation. Ten algorithms for CT data and eleven algorithms for MRI data, submitted from twelvegroups, have been evaluated. The results showed that the performance of CT WHS was generally better than thatof MRI WHS. The segmentation of the substructures for different categories of patients could present different levelsof challenge due to the difference in imaging and variations of heart shapes. The deep learning (DL)-based methodsdemonstrated great potential, though several of them reported poor results in the blinded evaluation. Their performance could vary greatly across different network structures and training strategies. The conventional algorithms,mainly based on multi-atlas segmentation, demonstrated good performance, though the accuracy and computationalefficiency could be limited. The challenge, including provision of the annotated training data and the blinded evaluation for submitted algorithms on the test data, conti
-
Journal articlePeghaire C, Dufton N, Lang M, et al., 2019,
The transcription factor ERG regulates a low shear stress-induced anti-thrombotic pathway in the microvasculature
, Nature Communications, Vol: 10, Pages: 1-17, ISSN: 2041-1723Endothelial cells actively maintain an anti-thrombotic environment; loss of this protective function may lead to thrombosis and systemic coagulopathy. The transcription factor ERG is essential to maintain endothelial homeostasis. Here we show that inducible endothelial ERG deletion (ErgiEC-KO) in mice is associated with spontaneous thrombosis, hemorrhages and systemic coagulopathy. We find that ERG drives transcription of the anti-coagulant thrombomodulin (TM), as shown by reporter assays and chromatin immunoprecipitation. TM expression is regulated by shear stress (SS) via Krüppel-like factor 2 (KLF2). In vitro, ERG regulates TM expression under low SS conditions, by facilitating KLF2 binding to the TM promoter. However, ERG is dispensable for TM expression in high SS conditions. In ErgiEC-KO mice, TM expression is decreased in liver and lung microvasculature exposed to low SS but not in blood vessels exposed to high SS. Our study identifies an endogenous, vascular bed- specific anti-coagulant pathway in microvasculature exposed to low SS.
-
Journal articleHultman K, Edsfeldt A, Björkbacka H, et al., 2019,
Cartilage oligomeric matrix protein associates with a vulnerable plaque phenotype in human atherosclerotic plaques
, Stroke, Vol: 50, ISSN: 0039-2499Background and Purpose- Extracellular matrix proteins are important in atherosclerotic disease by influencing plaque stability and cellular behavior but also by regulating inflammation. COMP (cartilage oligomeric matrix protein) is present in healthy human arteries and expressed by smooth muscle cells. A recent study showed that transplantation of COMP-deficient bone marrow to apoE-/- mice increased atherosclerotic plaque formation, indicating a role for COMP also in bone marrow-derived cells. Despite the evidence of a role for COMP in murine atherosclerosis, knowledge is lacking about the role of COMP in human atherosclerotic disease. Methods- In the present study, we investigated if COMP was associated with a stable or a vulnerable human atherosclerotic plaque phenotype by analyzing 211 carotid plaques for COMP expression using immunohistochemistry. Results- Plaque area that stained positive for COMP was significantly larger in atherosclerotic plaques associated with symptoms (n=110) compared with asymptomatic plaques (n=101; 9.7% [4.7-14.3] versus 5.6% [2.8-9.8]; P=0.0002). COMP was positively associated with plaque lipids (r=0.32; P=0.000002) and CD68 cells (r=0.15; P=0.036) but was negatively associated with collagen (r=-0.16; P=0.024), elastin (r=-0.14; P=0.041), and smooth muscle cells (r=-0.25; P=0.0002). COMP was positively associated with CD163 (r=0.37; P=0.00000006), a scavenger receptor for hemoglobin/haptoglobin and a marker of Mhem macrophages, and with intraplaque hemorrhage, measured as glycophorin A staining (r=0.28; P=0.00006). Conclusions- The present study shows that COMP is associated to symptomatic carotid atherosclerosis, CD163-expressing cells, and a vulnerable atherosclerotic plaque phenotype in humans.
-
Conference paperWang C, Papanastasiou G, Tsaftaris S, et al., 2019,
TPSDicyc: Improved deformation invariant cross-domain medical image synthesis
, Second International Workshop, MLMIR 2019, Held in Conjunction with MICCAI 2019, Publisher: Springer International Publishing, Pages: 245-254, ISSN: 0302-9743Cycle-consistent generative adversarial network (CycleGAN) has been widely used for cross-domain medical image systhesis tasks particularly due to its ability to deal with unpaired data. However, most CycleGAN-based synthesis methods can not achieve good alignment between the synthesized images and data from the source domain, even with additional image alignment losses. This is because the CycleGAN generator network can encode the relative deformations and noises associated to different domains. This can be detrimental for the downstream applications that rely on the synthesized images, such as generating pseudo-CT for PET-MR attenuation correction. In this paper, we present a deformation invariant model based on the deformation-invariant CycleGAN (DicycleGAN) architecture and the spatial transformation network (STN) using thin-plate-spline (TPS). The proposed method can be trained with unpaired and unaligned data, and generate synthesised images aligned with the source data. Robustness to the presence of relative deformations between data from the source and target domain has been evaluated through experiments on multi-sequence brain MR data and multi-modality abdominal CT and MR data. Experiment results demonstrated that our method can achieve better alignment between the source and target data while maintaining superior image quality of signal compared to several state-of-the-art CycleGAN-based methods.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.