1. |
- Underwood, J, et al.
(author)
-
Validation of a Novel Multivariate Method of Defining HIV-Associated Cognitive Impairment
- 2019
-
In: Open forum infectious diseases. - : Oxford University Press (OUP). - 2328-8957. ; 6:6, s. ofz198-
-
Journal article (peer-reviewed)abstract
- BackgroundThe optimum method of defining cognitive impairment in virally suppressed people living with HIV is unknown. We evaluated the relationships between cognitive impairment, including using a novel multivariate method (NMM), patient– reported outcome measures (PROMs), and neuroimaging markers of brain structure across 3 cohorts.MethodsDifferences in the prevalence of cognitive impairment, PROMs, and neuroimaging data from the COBRA, CHARTER, and POPPY cohorts (total n = 908) were determined between HIV-positive participants with and without cognitive impairment defined using the HIV-associated neurocognitive disorders (HAND), global deficit score (GDS), and NMM criteria.ResultsThe prevalence of cognitive impairment varied by up to 27% between methods used to define impairment (eg, 48% for HAND vs 21% for NMM in the CHARTER study). Associations between objective cognitive impairment and subjective cognitive complaints generally were weak. Physical and mental health summary scores (SF-36) were lowest for NMM-defined impairment (P < .05).There were no differences in brain volumes or cortical thickness between participants with and without cognitive impairment defined using the HAND and GDS measures. In contrast, those identified with cognitive impairment by the NMM had reduced mean cortical thickness in both hemispheres (P < .05), as well as smaller brain volumes (P < .01). The associations with measures of white matter microstructure and brain-predicted age generally were weaker.ConclusionDifferent methods of defining cognitive impairment identify different people with varying symptomatology and measures of brain injury. Overall, NMM-defined impairment was associated with most neuroimaging abnormalities and poorer self-reported health status. This may be due to the statistical advantage of using a multivariate approach.
|
|
2. |
- Beck, J. J., et al.
(author)
-
Genetic meta-analysis of twin birth weight shows high genetic correlation with singleton birth weight
- 2021
-
In: Human Molecular Genetics. - : Oxford University Press (OUP). - 0964-6906 .- 1460-2083. ; 30:19, s. 1894-1905
-
Journal article (peer-reviewed)abstract
- Birth weight (BW) is an important predictor of newborn survival and health and has associations with many adult health outcomes, including cardiometabolic disorders, autoimmune diseases and mental health. On average, twins have a lower BW than singletons as a result of a different pattern of fetal growth and shorter gestational duration. Therefore, investigations into the genetics of BW often exclude data from twins, leading to a reduction in sample size and remaining ambiguities concerning the genetic contribution to BW in twins. In this study, we carried out a genome-wide association meta-analysis of BW in 42 212 twin individuals and found a positive correlation of beta values (Pearson's r = 0.66, 95% confidence interval [CI]: 0.47-0.77) with 150 previously reported genome-wide significant variants for singleton BW. We identified strong positive genetic correlations between BW in twins and numerous anthropometric traits, most notably with BW in singletons (genetic correlation [r(g)]= 0.92, 95% CI: 0.66-1.18). Genetic correlations of BW in twins with a series of health-related traits closely resembled those previously observed for BW in singletons. Polygenic scores constructed from a genome-wide association study on BW in the UK Biobank demonstrated strong predictive power in a target sample of Dutch twins and singletons. Together, our results indicate that a similar genetic architecture underlies BW in twins and singletons and that future genome-wide studies might benefit from including data from large twin registers.
|
|
3. |
|
|
4. |
- Hammar, Mats, et al.
(author)
-
Tibolone and low-dose continuous combined hormone treatment : Vaginal bleeding pattern, efficacy and tolerability
- 2007
-
In: British Journal of Obstetrics and Gynecology. - : Wiley. - 1470-0328 .- 1471-0528. ; 114:12, s. 1522-1529
-
Journal article (peer-reviewed)abstract
- Objectives: The primary objective was to compare the vaginal bleeding pattern during administration of tibolone and low-dose continuous combined estradiol plus norethisterone acetate (E2/NETA). The secondary objectives were efficacy on vasomotor symptoms and vaginal atrophy. Design: A randomised, double-blind, double-dummy, group comparative intervention trial. Setting: Multicentre study executed in 32 centres in 7 European countries. Sample: Five hundred and seventy-two healthy symptomatic postmenopausal women, aged 45-65 years. Methods: Participants were randomised to receive 2.5 mg tibolone or 1 mg 17ß estradiol plus 0.5 mg norethisterone acetate (E 2/NETA) daily for 48 weeks. Main outcome measures: Prevalence of vaginal bleeding, hot flushes and adverse events. Results: The incidence of bleeding was significantly lower in the tibolone group during the first 3 months of treatment (18.3 versus 33.1%, P < 0.001) when compared with the E 2/NETA group. This effect on the bleeding pattern was sustained throughout the study, although reaching statistical significance again only in 7-9 months of treatment (11 versus 19%, P < 0.05). In both treatment groups, vasomotor symptoms and vaginal atrophy were significantly reduced to a similar extent when compared with baseline. The prevalence of breast pain/tenderness was significantly lower with tibolone compared with E2/NETA (3.2 versus 9.8%, P < 0.001). Conclusion: Tibolone reduces menopausal symptoms to a similar extent as conventional low-dose continuous combined hormone therapy but causes significant less vaginal bleeding in the first 3 months of treatment. This constitutes an important argument for woman adherence to therapy. © 2007 The Authors.
|
|
5. |
- Khan, Fahad, et al.
(author)
-
Scale coding bag-of-words for action recognition
- 2014
-
In: Pattern Recognition (ICPR), 2014 22nd International Conference on. - : Institute of Electrical and Electronics Engineers Inc.. - 9781479952083 ; , s. 1514-1519
-
Conference paper (peer-reviewed)abstract
- Recognizing human actions in still images is a challenging problem in computer vision due to significant amount of scale, illumination and pose variation. Given the bounding box of a person both at training and test time, the task is to classify the action associated with each bounding box in an image. Most state-of-the-art methods use the bag-of-words paradigm for action recognition. The bag-of-words framework employing a dense multi-scale grid sampling strategy is the de facto standard for feature detection. This results in a scale invariant image representation where all the features at multiple-scales are binned in a single histogram. We argue that such a scale invariant strategy is sub-optimal since it ignores the multi-scale information available with each bounding box of a person. This paper investigates alternative approaches to scale coding for action recognition in still images. We encode multi-scale information explicitly in three different histograms for small, medium and large scale visual-words. Our first approach exploits multi-scale information with respect to the image size. In our second approach, we encode multi-scale information relative to the size of the bounding box of a person instance. In each approach, the multi-scale histograms are then concatenated into a single representation for action classification. We validate our approaches on the Willow dataset which contains seven action categories: interacting with computer, photography, playing music, riding bike, riding horse, running and walking. Our results clearly suggest that the proposed scale coding approaches outperform the conventional scale invariant technique. Moreover, we show that our approach obtains promising results compared to more complex state-of-the-art methods.
|
|
6. |
- Reijmers, Rogier M., et al.
(author)
-
Disruption of heparan sulfate proteoglycan conformation perturbs B-cell maturation and APRIL-mediated plasma cell survival
- 2011
-
In: Blood. - : American Society of Hematology. - 0006-4971 .- 1528-0020. ; 117:23, s. 6162-6171
-
Journal article (peer-reviewed)abstract
- The development and antigen-dependent differentiation of B lymphocytes are orchestrated by an array of growth factors, cytokines, and chemokines that require tight spatiotemporal regulation. Heparan sulfate proteoglycans specifically bind and regulate the bioavailability of soluble protein ligands, but their role in the immune system has remained largely unexplored. Modification of heparan sulfate by glucuronyl C5-epimerase (Glce) controls heparan sulfate-chain flexibility and thereby affects ligand binding. Here we show that Glce deficiency impairs B-cell maturation, resulting in decreased plasma cell numbers and immunoglobulin levels. We demonstrate that C5-epimerase modification of heparan sulfate is critical for binding of a proliferation inducing ligand (APRIL) and that Glce-deficient plasma cells fail to respond to APRIL-mediated survival signals. Our results identify heparan sulfate proteoglycans as novel players in B-cell maturation and differentiation and suggest that heparan sulfate conformation is crucial for recruitment of factors that control plasma cell survival.
|
|
7. |
- Wang, Y., et al.
(author)
-
MineGAN : Effective Knowledge Transfer From GANs to Target Domains With Few Images
- 2020
-
In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). - : IEEE. - 9781728171685 ; , s. 9329-9338
-
Conference paper (peer-reviewed)abstract
- One of the attractive characteristics of deep neural networks is their ability to transfer knowledge obtained in one domain to other related domains. As a result, high-quality networks can be trained in domains with relatively little training data. This property has been extensively studied for discriminative networks but has received significantly less attention for generative models. Given the often enormous effort required to train GANs, both computationally as well as in the dataset collection, the re-use of pretrained GANs is a desirable objective. We propose a novel knowledge transfer method for generative models based on mining the knowledge that is most beneficial to a specific target domain, either from a single or multiple pretrained GANs. This is done using a miner network that identifies which part of the generative distribution of each pretrained GAN outputs samples closest to the target domain. Mining effectively steers GAN sampling towards suitable regions of the latent space, which facilitates the posterior finetuning and avoids pathologies of other methods such as mode collapse and lack of flexibility. We perform experiments on several complex datasets using various GAN architectures (BigGAN, Progressive GAN) and show that the proposed method, called MineGAN, effectively transfers knowledge to domains with few target images, outperforming existing methods. In addition, MineGAN can successfully transfer knowledge from multiple pretrained GANs. Our code is available at: https://github.com/yaxingwang/MineGAN.
|
|
8. |
- Wang, Y., et al.
(author)
-
Semi-Supervised Learning for Few-Shot Image-to-Image Translation
- 2020
-
In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). - : IEEE. - 9781728171685 ; , s. 4452-4461
-
Conference paper (peer-reviewed)abstract
- In the last few years, unpaired image-to-image translation has witnessed Remarkable progress. Although the latest methods are able to generate realistic images, they crucially rely on a large number of labeled images. Recently, some methods have tackled the challenging setting of few-shot image-to-image ranslation, reducing the labeled data requirements for the target domain during inference. In this work, we go one step further and reduce the amount of required labeled data also from the source domain during training. To do so, we propose applying semi-supervised learning via a noise-tolerant pseudo-labeling procedure. We also apply a cycle consistency constraint to further exploit the information from unlabeled images, either from the same dataset or external. Additionally, we propose several structural modifications to facilitate the image translation task under these circumstances. Our semi-supervised method for few-shot image translation, called SEMIT, achieves excellent results on four different datasets using as little as 10% of the source labels, and matches the performance of the main fully-supervised competitor using only 20% labeled data. Our code and models are made public at: https://github.com/yaxingwang/SEMIT.
|
|
9. |
- Zlatev, Jordan, et al.
(author)
-
Idiomatic entrenchment and semantic priming
- 2004
-
In: Linguagem, Cultura e Cognição. - Coimbra : Almedina. - 9724022641 - 9724023036 ; , s. 309-334
-
Book chapter (other academic/artistic)
|
|