| 1. |
- Arking, D. E., et al.
(författare)
-
Genetic association study of QT interval highlights role for calcium signaling pathways in myocardial repolarization
- 2014
-
Ingår i: Nature Genetics. - : Nature Publishing Group. - 1061-4036 .- 1546-1718. ; 46:8, s. 826-836
-
Tidskriftsartikel (refereegranskat)abstract
- The QT interval, an electrocardiographic measure reflecting myocardial repolarization, is a heritable trait. QT prolongation is a risk factor for ventricular arrhythmias and sudden cardiac death (SCD) and could indicate the presence of the potentially lethal mendelian long-QT syndrome (LQTS). Using a genome-wide association and replication study in up to 100,000 individuals, we identified 35 common variant loci associated with QT interval that collectively explain ∼ 8-10% of QT-interval variation and highlight the importance of calcium regulation in myocardial repolarization. Rare variant analysis of 6 new QT interval-associated loci in 298 unrelated probands with LQTS identified coding variants not found in controls but of uncertain causality and therefore requiring validation. Several newly identified loci encode proteins that physically interact with other recognized repolarization proteins. Our integration of common variant association, expression and orthogonal protein-protein interaction screens provides new insights into cardiac electrophysiology and identifies new candidate genes for ventricular arrhythmias, LQTS and SCD. © 2014 Nature America, Inc.
|
|
| 2. |
- Shungin, Dmitry, et al.
(författare)
-
New genetic loci link adipose and insulin biology to body fat distribution.
- 2015
-
Ingår i: Nature. - : Springer Science and Business Media LLC. - 0028-0836 .- 1476-4687. ; 518:7538, s. 187-378
-
Tidskriftsartikel (refereegranskat)abstract
- Body fat distribution is a heritable trait and a well-established predictor of adverse metabolic outcomes, independent of overall adiposity. To increase our understanding of the genetic basis of body fat distribution and its molecular links to cardiometabolic traits, here we conduct genome-wide association meta-analyses of traits related to waist and hip circumferences in up to 224,459 individuals. We identify 49 loci (33 new) associated with waist-to-hip ratio adjusted for body mass index (BMI), and an additional 19 loci newly associated with related waist and hip circumference measures (P < 5 × 10(-8)). In total, 20 of the 49 waist-to-hip ratio adjusted for BMI loci show significant sexual dimorphism, 19 of which display a stronger effect in women. The identified loci were enriched for genes expressed in adipose tissue and for putative regulatory elements in adipocytes. Pathway analyses implicated adipogenesis, angiogenesis, transcriptional regulation and insulin resistance as processes affecting fat distribution, providing insight into potential pathophysiological mechanisms.
|
|
| 3. |
- Locke, Adam E, et al.
(författare)
-
Genetic studies of body mass index yield new insights for obesity biology.
- 2015
-
Ingår i: Nature. - : Springer Science and Business Media LLC. - 0028-0836 .- 1476-4687. ; 518:7538, s. 197-401
-
Tidskriftsartikel (refereegranskat)abstract
- Obesity is heritable and predisposes to many diseases. To understand the genetic basis of obesity better, here we conduct a genome-wide association study and Metabochip meta-analysis of body mass index (BMI), a measure commonly used to define obesity and assess adiposity, in up to 339,224 individuals. This analysis identifies 97 BMI-associated loci (P < 5 × 10(-8)), 56 of which are novel. Five loci demonstrate clear evidence of several independent association signals, and many loci have significant effects on other metabolic phenotypes. The 97 loci account for ∼2.7% of BMI variation, and genome-wide estimates suggest that common variation accounts for >20% of BMI variation. Pathway analyses provide strong support for a role of the central nervous system in obesity susceptibility and implicate new genes and pathways, including those related to synaptic function, glutamate signalling, insulin secretion/action, energy metabolism, lipid biology and adipogenesis.
|
|
| 4. |
- Kanoni, Stavroula, et al.
(författare)
-
Implicating genes, pleiotropy, and sexual dimorphism at blood lipid loci through multi-ancestry meta-analysis.
- 2022
-
Ingår i: Genome biology. - : Springer Science and Business Media LLC. - 1474-760X .- 1465-6906 .- 1474-7596. ; 23:1
-
Tidskriftsartikel (refereegranskat)abstract
- Genetic variants within nearly 1000 loci are known to contribute to modulation of blood lipid levels. However, the biological pathways underlying these associations are frequently unknown, limiting understanding of these findings and hindering downstream translational efforts such as drug target discovery.To expand our understanding of the underlying biological pathways and mechanisms controlling blood lipid levels, we leverage a large multi-ancestry meta-analysis (N=1,654,960) of blood lipids to prioritize putative causal genes for 2286 lipid associations using six gene prediction approaches. Using phenome-wide association (PheWAS) scans, we identify relationships of genetically predicted lipid levels to other diseases and conditions. We confirm known pleiotropic associations with cardiovascular phenotypes and determine novel associations, notably with cholelithiasis risk. We perform sex-stratified GWAS meta-analysis of lipid levels and show that 3-5% of autosomal lipid-associated loci demonstrate sex-biased effects. Finally, we report 21 novel lipid loci identified on the X chromosome. Many of the sex-biased autosomal and X chromosome lipid loci show pleiotropic associations with sex hormones, emphasizing the role of hormone regulation in lipid metabolism.Taken together, our findings provide insights into the biological mechanisms through which associated variants lead to altered lipid levels and potentially cardiovascular disease risk.
|
|
| 5. |
- Justice, Anne E., et al.
(författare)
-
Protein-coding variants implicate novel genes related to lipid homeostasis contributing to body-fat distribution
- 2019
-
Ingår i: Nature Genetics. - : Nature Publishing Group. - 1061-4036 .- 1546-1718. ; 51:3, s. 452-469
-
Tidskriftsartikel (refereegranskat)abstract
- Body-fat distribution is a risk factor for adverse cardiovascular health consequences. We analyzed the association of body-fat distribution, assessed by waist-to-hip ratio adjusted for body mass index, with 228,985 predicted coding and splice site variants available on exome arrays in up to 344,369 individuals from five major ancestries (discovery) and 132,177 European-ancestry individuals (validation). We identified 15 common (minor allele frequency, MAF >= 5%) and nine low-frequency or rare (MAF < 5%) coding novel variants. Pathway/gene set enrichment analyses identified lipid particle, adiponectin, abnormal white adipose tissue physiology and bone development and morphology as important contributors to fat distribution, while cross-trait associations highlight cardiometabolic traits. In functional follow-up analyses, specifically in Drosophila RNAi-knockdowns, we observed a significant increase in the total body triglyceride levels for two genes (DNAH10 and PLXND1). We implicate novel genes in fat distribution, stressing the importance of interrogating low-frequency and protein-coding variants.
|
|
| 6. |
- Marouli, Eirini, et al.
(författare)
-
Rare and low-frequency coding variants alter human adult height
- 2017
-
Ingår i: Nature. - : Springer Science and Business Media LLC. - 0028-0836 .- 1476-4687. ; 542:7640, s. 186-190
-
Tidskriftsartikel (refereegranskat)abstract
- Height is a highly heritable, classic polygenic trait with approximately 700 common associated variants identified through genome-wide association studies so far. Here, we report 83 height-associated coding variants with lower minor-allele frequencies (in the range of 0.1-4.8%) and effects of up to 2 centimetres per allele (such as those in IHH, STC2, AR and CRISPLD2), greater than ten times the average effect of common variants. In functional follow-up studies, rare height increasing alleles of STC2 (giving an increase of 1-2 centimetres per allele) compromised proteolytic inhibition of PAPP-A and increased cleavage of IGFBP-4 in vitro, resulting in higher bioavailability of insulin-like growth factors. These 83 height-associated variants overlap genes that are mutated in monogenic growth disorders and highlight new biological candidates (such as ADAMTS3, IL11RA and NOX4) and pathways (such as proteoglycan and glycosaminoglycan synthesis) involved in growth. Our results demonstrate that sufficiently large sample sizes can uncover rare and low-frequency variants of moderate-to-large effect associated with polygenic human phenotypes, and that these variants implicate relevant genes and pathways.
|
|
| 7. |
- Speliotes, Elizabeth K., et al.
(författare)
-
Association analyses of 249,796 individuals reveal 18 new loci associated with body mass index
- 2010
-
Ingår i: Nature Genetics. - : Springer Science and Business Media LLC. - 1061-4036 .- 1546-1718. ; 42:11, s. 937-948
-
Tidskriftsartikel (refereegranskat)abstract
- Obesity is globally prevalent and highly heritable, but its underlying genetic factors remain largely elusive. To identify genetic loci for obesity susceptibility, we examined associations between body mass index and ~2.8 million SNPs in up to 123,865 individuals with targeted follow up of 42 SNPs in up to 125,931 additional individuals. We confirmed 14 known obesity susceptibility loci and identified 18 new loci associated with body mass index (P < 5 × 10−8), one of which includes a copy number variant near GPRC5B. Some loci (at MC4R, POMC, SH2B1 and BDNF) map near key hypothalamic regulators of energy balance, and one of these loci is near GIPR, an incretin receptor. Furthermore, genes in other newly associated loci may provide new insights into human body weight regulation.
|
|
| 8. |
- Berndt, Sonja I., et al.
(författare)
-
Genome-wide meta-analysis identifies 11 new loci for anthropometric traits and provides insights into genetic architecture
- 2013
-
Ingår i: Nature Genetics. - : Springer Science and Business Media LLC. - 1061-4036 .- 1546-1718. ; 45:5, s. 501-U69
-
Tidskriftsartikel (refereegranskat)abstract
- Approaches exploiting trait distribution extremes may be used to identify loci associated with common traits, but it is unknown whether these loci are generalizable to the broader population. In a genome-wide search for loci associated with the upper versus the lower 5th percentiles of body mass index, height and waist-to-hip ratio, as well as clinical classes of obesity, including up to 263,407 individuals of European ancestry, we identified 4 new loci (IGFBP4, H6PD, RSRC1 and PPP2R2A) influencing height detected in the distribution tails and 7 new loci (HNF4G, RPTOR, GNAT2, MRPS33P4, ADCY9, HS6ST3 and ZZZ3) for clinical classes of obesity. Further, we find a large overlap in genetic structure and the distribution of variants between traits based on extremes and the general population and little etiological heterogeneity between obesity subgroups.
|
|
| 9. |
- Lu, Yingchang, et al.
(författare)
-
New loci for body fat percentage reveal link between adiposity and cardiometabolic disease risk
- 2016
-
Ingår i: Nature Communications. - : Springer Science and Business Media LLC. - 2041-1723. ; 7
-
Tidskriftsartikel (refereegranskat)abstract
- To increase our understanding of the genetic basis of adiposity and its links to cardiometabolic disease risk, we conducted a genome-wide association meta-analysis of body fat percentage (BF%) in up to 100,716 individuals. Twelve loci reached genome-wide significance (P<5 × 10(-8)), of which eight were previously associated with increased overall adiposity (BMI, BF%) and four (in or near COBLL1/GRB14, IGF2BP1, PLA2G6, CRTC1) were novel associations with BF%. Seven loci showed a larger effect on BF% than on BMI, suggestive of a primary association with adiposity, while five loci showed larger effects on BMI than on BF%, suggesting association with both fat and lean mass. In particular, the loci more strongly associated with BF% showed distinct cross-phenotype association signatures with a range of cardiometabolic traits revealing new insights in the link between adiposity and disease risk.
|
|
| 10. |
- Turcot, Valerie, et al.
(författare)
-
Protein-altering variants associated with body mass index implicate pathways that control energy intake and expenditure in obesity
- 2018
-
Ingår i: Nature Genetics. - : Nature Publishing Group. - 1061-4036 .- 1546-1718. ; 50:1, s. 26-41
-
Tidskriftsartikel (refereegranskat)abstract
- Genome-wide association studies (GWAS) have identified >250 loci for body mass index (BMI), implicating pathways related to neuronal biology. Most GWAS loci represent clusters of common, noncoding variants from which pinpointing causal genes remains challenging. Here we combined data from 718,734 individuals to discover rare and low-frequency (minor allele frequency (MAF) < 5%) coding variants associated with BMI. We identified 14 coding variants in 13 genes, of which 8 variants were in genes (ZBTB7B, ACHE, RAPGEF3, RAB21, ZFHX3, ENTPD6, ZFR2 and ZNF169) newly implicated in human obesity, 2 variants were in genes (MC4R and KSR2) previously observed to be mutated in extreme obesity and 2 variants were in GIPR. The effect sizes of rare variants are similar to 10 times larger than those of common variants, with the largest effect observed in carriers of an MC4R mutation introducing a stop codon (p.Tyr35Ter, MAF = 0.01%), who weighed similar to 7 kg more than non-carriers. Pathway analyses based on the variants associated with BMI confirm enrichment of neuronal genes and provide new evidence for adipocyte and energy expenditure biology, widening the potential of genetically supported therapeutic targets in obesity.
|
|
| 11. |
|
|
| 12. |
- Ancuti, Codruta O., et al.
(författare)
-
NTIRE 2023 HR NonHomogeneous Dehazing Challenge Report
- 2023
-
Ingår i: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). - Vancover : Institute of Electrical and Electronics Engineers (IEEE).
-
Konferensbidrag (refereegranskat)abstract
- This study assesses the outcomes of the NTIRE 2023 Challenge on Non-Homogeneous Dehazing, wherein novel techniques were proposed and evaluated on new image dataset called HD-NH-HAZE. The HD-NH-HAZE dataset contains 50 high resolution pairs of real-life outdoor images featuring nonhomogeneous hazy images and corresponding haze-free images of the same scene. The nonhomogeneous haze was simulated using a professional setup that replicated real-world conditions of hazy scenarios. The competition had 246 participants and 17 teams that competed in the final testing phase, and the proposed solutions demonstrated the cutting-edge in image dehazing technology.
|
|
| 13. |
- Back, A., et al.
(författare)
-
Complexity metric based on fraction of penumbra dose - initial study
- 2017
-
Ingår i: 9TH INTERNATIONAL CONFERENCE ON 3D RADIATION DOSIMETRY ; Journal of Physics Conference Series. - : IOP Publishing. - 1742-6588 .- 1742-6596.
-
Bokkapitel (övrigt vetenskapligt/konstnärligt)abstract
- Volumetric modulated arc therapy improve radiotherapy outcome for many patients compared to conventional three dimensional conformal radiotherapy but require a more extensive, most often measurement based, quality assurance. Multi leaf collimator (MLC) aperture-based complexity metrics have been suggested to be used to distinguish complex treatment plans unsuitable for treatment without time consuming measurements. This study introduce a spatially resolved complexity score that correlate to the fraction of penumbra dose and will give information on the spatial distribution and the clinical relevance of the calculated complexity. The complexity metric is described and an initial study on the correlation between the complexity score and the difference between measured and calculated dose for 30 MLC openings is presented. The result of an analysis of the complexity scores were found to correlate to differences between measurements and calculations with a Pearson's r-value of 0.97.
|
|
| 14. |
- Bruhn, Sören, et al.
(författare)
-
A Generally Applicable Translational Strategy Identifies S100A4 as a Candidate Gene in Allergy
- 2014
-
Ingår i: Science Translational Medicine. - : American Association for the Advancement of Science (AAAS). - 1946-6234 .- 1946-6242. ; 6:218
-
Tidskriftsartikel (refereegranskat)abstract
- The identification of diagnostic markers and therapeutic candidate genes in common diseases is complicated by the involvement of thousands of genes. We hypothesized that genes co-regulated with a key gene in allergy, IL13, would form a module that could help to identify candidate genes. We identified a T helper 2 (T(H)2) cell module by small interfering RNA-mediated knockdown of 25 putative IL13-regulating transcription factors followed by expression profiling. The module contained candidate genes whose diagnostic potential was supported by clinical studies. Functional studies of human TH2 cells as well as mouse models of allergy showed that deletion of one of the genes, S100A4, resulted in decreased signs of allergy including TH2 cell activation, humoral immunity, and infiltration of effector cells. Specifically, dendritic cells required S100A4 for activating T cells. Treatment with an anti-S100A4 antibody resulted in decreased signs of allergy in the mouse model as well as in allergen-challenged T cells from allergic patients. This strategy, which may be generally applicable to complex diseases, identified and validated an important diagnostic and therapeutic candidate gene in allergy.
|
|
| 15. |
- Conde, Marcus V., et al.
(författare)
-
Lens-to-Lens Bokeh Effect Transformation : NTIRE 2023 Challenge Report
- 2023
-
Ingår i: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. - Vancover : Institute of Electrical and Electronics Engineers (IEEE). ; , s. 1643-1659
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- We present the new Bokeh Effect Transformation Dataset (BETD), and review the proposed solutions for this novel task at the NTIRE 2023 Bokeh Effect Transformation Challenge. Recent advancements of mobile photography aim to reach the visual quality of full-frame cameras. Now, a goal in computational photography is to optimize the Bokeh effect itself, which is the aesthetic quality of the blur in out-of-focus areas of an image. Photographers create this aesthetic effect by benefiting from the lens optical properties. The aim of this work is to design a neural network capable of converting the the Bokeh effect of one lens to the effect of another lens without harming the sharp foreground regions in the image. For a given input image, knowing the target lens type, we render or transform the Bokeh effect accordingly to the lens properties. We build the BETD using two full-frame Sony cameras, and diverse lens setups. To the best of our knowledge, we are the first attempt to solve this novel task, and we provide the first BETD dataset and benchmark for it. The challenge had 99 registered participants. The submitted methods gauge the state-of-the-art in Bokeh effect rendering and transformation.
|
|
| 16. |
- Fullagar, Wilfred K., et al.
(författare)
-
Beating Darwin-Bragg losses in lab-based ultrafast X-ray experiments
- 2017
-
Ingår i: Structural Dynamics. - : AIP Publishing. - 2329-7778. ; 4:4
-
Tidskriftsartikel (refereegranskat)abstract
- The use of low temperature thermal detectors for avoiding Darwin-Bragg losses in lab-based ultrafast experiments has begun. An outline of the background of this new development is offered, showing the relevant history and initiative taken by this work.
|
|
| 17. |
|
|
| 18. |
|
|
| 19. |
- Gustafsson, Fredrik K., et al.
(författare)
-
Accurate 3D Object Detection using Energy-Based Models
- 2021
-
Ingår i: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recogition Workshops (CVPRW 2021). - : Institute of Electrical and Electronics Engineers (IEEE). - 9781665448994 ; , s. 2849-2858
-
Konferensbidrag (refereegranskat)abstract
- Accurate 3D object detection (3DOD) is crucial for safe navigation of complex environments by autonomous robots. Regressing accurate 3D bounding boxes in cluttered environments based on sparse LiDAR data is however a highly challenging problem. We address this task by exploring recent advances in conditional energy-based models (EBMs) for probabilistic regression. While methods employing EBMs for regression have demonstrated impressive performance on 2D object detection in images, these techniques are not directly applicable to 3D bounding boxes. In this work, we therefore design a differentiable pooling operator for 3D bounding boxes, serving as the core module of our EBM network. We further integrate this general approach into the state-of-the-art 3D object detector SA-SSD. On the KITTI dataset, our proposed approach consistently outperforms the SA-SSD baseline across all 3DOD metrics, demonstrating the potential of EBM-based regression for highly accurate 3DOD. Code is available at https://github.com/fregu856/ebms_3dod.
|
|
| 20. |
|
|
| 21. |
- Gustafsson, Fredrik K., et al.
(författare)
-
Evaluating Scalable Bayesian Deep Learning Methods for Robust Computer Vision
- 2020
-
Ingår i: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW 2020). - : IEEE Computer Society. - 9781728193601 ; , s. 1289-1298
-
Konferensbidrag (refereegranskat)abstract
- While deep neural networks have become the go-to approach in computer vision, the vast majority of these models fail to properly capture the uncertainty inherent in their predictions. Estimating this predictive uncertainty can be crucial, for example in automotive applications. In Bayesian deep learning, predictive uncertainty is commonly decomposed into the distinct types of aleatoric and epistemic uncertainty. The former can be estimated by letting a neural network output the parameters of a certain probability distribution. Epistemic uncertainty estimation is a more challenging problem, and while different scalable methods recently have emerged, no extensive comparison has been performed in a real-world setting. We therefore accept this task and propose a comprehensive evaluation framework for scalable epistemic uncertainty estimation methods in deep learning. Our proposed framework is specifically designed to test the robustness required in real-world computer vision applications. We also apply this framework to provide the first properly extensive and conclusive comparison of the two current state-of-the-art scalable methods: ensembling and MC-dropout. Our comparison demonstrates that ensembling consistently provides more reliable and practically useful uncertainty estimates. Code is available at https://github.com/fregu856/evaluating_bdl.
|
|
| 22. |
|
|
| 23. |
|
|
| 24. |
- Gustafsson, Fredrik K., et al.
(författare)
-
Learning Proposals for Practical Energy-Based Regression
- 2022
-
Ingår i: International conference on artificial intelligence and statistics, vol 151. - : JMLR-JOURNAL MACHINE LEARNING RESEARCH. ; , s. 4685-4704
-
Konferensbidrag (refereegranskat)abstract
- Energy-based models (EBMs) have experienced a resurgence within machine learning in recent years, including as a promising alternative for probabilistic regression. However, energy-based regression requires a proposal distribution to be manually designed for training, and an initial estimate has to be provided at test-time. We address both of these issues by introducing a conceptually simple method to automatically learn an effective proposal distribution, which is parameterized by a separate network head. To this end, we derive a surprising result, leading to a unified training objective that jointly minimizes the KL divergence from the proposal to the EBM, and the negative log-likelihood of the EBM. At test-time, we can then employ importance sampling with the trained proposal to efficiently evaluate the learned EBM and produce standalone predictions. Furthermore, we utilize our derived training objective to learn mixture density networks (MDNs) with a jointly trained energy-based teacher, consistently outperforming conventional MDN training on four real-world regression tasks within computer vision. Code is available at https://github.com/fregu856/ebms_proposals.
|
|
| 25. |
- Gustafsson, Fredrik K., 1993-
(författare)
-
Towards Accurate and Reliable Deep Regression Models
- 2023
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Regression is a fundamental machine learning task with many important applications within computer vision and other domains. In general, it entails predicting continuous targets from given inputs. Deep learning has become the dominant paradigm within machine learning in recent years, and a wide variety of different techniques have been employed to solve regression problems using deep models. There is however no broad consensus on how deep regression models should be constructed for best possible accuracy, or how the uncertainty in their predictions should be represented and estimated. These open questions are studied in this thesis, aiming to help take steps towards an ultimate goal of developing deep regression models which are both accurate and reliable enough for real-world deployment within medical applications and other safety-critical domains.The first main contribution of the thesis is the formulation and development of energy-based probabilistic regression. This is a general and conceptually simple regression framework with a clear probabilistic interpretation, using energy-based models to represent the true conditional target distribution. The framework is applied to a number of regression problems and demonstrates particularly strong performance for 2D bounding box regression, improving the state-of-the-art when applied to the task of visual tracking.The second main contribution is a critical evaluation of various uncertainty estimation methods. A general introduction to the problem of estimating the predictive uncertainty of deep models is first provided, together with an extensive comparison of the two popular methods ensembling and MC-dropout. A number of regression uncertainty estimation methods are then further evaluated, specifically examining their reliability under real-world distribution shifts. This evaluation uncovers important limitations of current methods and serves as a challenge to the research community. It demonstrates that more work is required in order to develop truly reliable uncertainty estimation methods for regression.
|
|
| 26. |
- Hendriks, Johannes N., et al.
(författare)
-
Deep Energy-Based NARX Models
- 2021
-
Ingår i: IFAC PapersOnLine. - : Elsevier. - 2405-8963. ; , s. 505-510
-
Konferensbidrag (refereegranskat)abstract
- This paper is directed towards the problem of learning nonlinear ARX models based on observed input output data. In particular, our interest is in learning a conditional distribution of the current output based on a finite window of past inputs and outputs. To achieve this, we consider the use of so-called energy-based models, which have been developed in allied fields for learning unknown distributions based on data. This energy-based model relies on a general function to describe the distribution, and here we consider a deep neural network for this purpose. The primary benefit of this approach is that it is capable of learning both simple and highly complex noise models, which we demonstrate on simulated and experimental data.
|
|
| 27. |
- Langner, Taro, et al.
(författare)
-
Uncertainty-Aware Body Composition Analysis with Deep Regression Ensembles on UK Biobank MRI
- 2021
-
Ingår i: Computerized Medical Imaging and Graphics. - : Elsevier BV. - 0895-6111 .- 1879-0771. ; 93
-
Tidskriftsartikel (refereegranskat)abstract
- Along with rich health-related metadata, an ongoing imaging study has acquired MRI of over 40,000 male and female UK Biobank participants aged 44-82 since 2014. Phenotypes derived from these images, such as measurements of body composition, can reveal new links between genetics, cardiovascular disease, and metabolic conditions. In this retrospective study, six measurements of body composition were automatically estimated by ResNet50 neural networks for image-based regression from neck-to-knee body MRI. Despite the potential for high speed and accuracy, these networks produce no output segmentations that could indicate the reliability of individual measurements. The presented experiments therefore examine mean-variance regression and ensembling for predictive uncertainty estimation, which can quantify individual measurement errors and thereby help to identify potential outliers, anomalies, and other failure cases automatically. In 10-fold cross-validation on data of about 8,500 subjects, mean-variance regression and ensembling showed complementary benefits, reducing the mean absolute error across all predictions by 12%. Both improved the calibration of uncertainties and their ability to identify high prediction errors. With intra-class correlation coefficients (ICC) above 0.97, all targets except the liver fat content yielded relative measurement errors below 5%. Testing on another 1,000 subjects showed consistent performance, and the method was finally deployed for inference to 30,000 subjects with missing reference values. The results indicate that deep regression ensembles could ultimately provide automated, uncertainty-aware measurements of body composition for more than 120,000 UK Biobank neck-to-knee body MRI that are to be acquired within the coming years.
|
|
| 28. |
- Luo, Ziwei, et al.
(författare)
-
Image Restoration with Mean-Reverting Stochastic Differential Equations
- 2023
-
Ingår i: Proceedings of the 40th International Conference on Machine Learning. ; , s. 23045-23066
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- This paper presents a stochastic differential equation (SDE) approach for general-purpose image restoration. The key construction consists in a mean-reverting SDE that transforms a high-quality image into a degraded counterpart as a mean state with fixed Gaussian noise. Then, by simulating the corresponding reverse-time SDE, we are able to restore the origin of the low-quality image without relying on any task-specific prior knowledge. Crucially, the proposed mean-reverting SDE has a closed-form solution, allowing us to compute the ground truth time-dependent score and learn it with a neural network. Moreover, we propose a maximum likelihood objective to learn an optimal reverse trajectory that stabilizes the training and improves the restoration results. The experiments show that our proposed method achieves highly competitive performance in quantitative comparisons on image deraining, deblurring, and denoising, setting a new state-of-the-art on two deraining datasets. Finally, the general applicability of our approach is further demonstrated via qualitative results on image super-resolution, inpainting, and dehazing. Code is available at https://github.com/Algolzw/image-restoration-sde.
|
|
| 29. |
- Luo, Ziwei, et al.
(författare)
-
Refusion : Enabling Large-Size Realistic Image Restoration With Latent-Space Diffusion Models
- 2023
-
Ingår i: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). - : Institute of Electrical and Electronics Engineers (IEEE). ; , s. 1680-1691
-
Konferensbidrag (refereegranskat)abstract
- This work aims to improve the applicability of diffusion models in realistic image restoration. Specifically, we enhance the diffusion model in several aspects such as network architecture, noise level, denoising steps, training image size, and optimizer/scheduler. We show that tuning these hyperparameters allows us to achieve better performance on both distortion and perceptual scores. We also propose a U-Net based latent diffusion model which performs diffusion in a low-resolution latent space while preserving high-resolution information from the original input for the decoding process. Compared to the previous latent-diffusion model which trains a VAE-GAN to compress the image, our proposed U-Net compression strategy is significantly more stable and can recover highly accurate images without relying on adversarial optimization. Importantly, these modifications allow us to apply diffusion models to various image restoration tasks, including real-world shadow removal, HR non-homogeneous dehazing, stereo super-resolution, and bokeh effect transformation. By simply replacing the datasets and slightly changing the noise network, our model, named Refusion, is able to deal with large-size images (e.g., 6000 x 4000 x 3 in HR dehazing) and produces good results on all the above restoration problems. Our Refusion achieves the best perceptual performance in the NTIRE 2023 Image Shadow Removal Challenge and wins 2nd place overall.
|
|
| 30. |
- Maercker, Matthias, 1979, et al.
(författare)
-
The detached dust and gas shells around the carbon star U Ant
- 2009
-
Tidskriftsartikel (övrigt vetenskapligt/konstnärligt)abstract
- Context: Geometrically thin, detached shells of gas have been found around a handful of carbon stars. Aims: Previous observations of scattered stellar light in the circumstellar medium around the carbon star U Ant were taken through filters centred on the resonance lines of K and Na. These observations could not separate the scattering by dust and atoms. The aim of this paper is to remedy this situation. Methods: We have obtained polarization data on stellar light scattered in the circumstellar medium around U Ant through filters which contain no strong lines, making it possible to differentiate between the two scattering agents. Kinematic, as well as spatial, information on the gas shells were obtained through high-resolution echelle spectrograph observations of the KI and NaD lines. Results: We confirm the existence of two detached shells around U Ant. The inner shell (at a radius of approx 43" and a width of approx 2") consists mainly of gas, while the outer shell (at a radius of approx 50" and a width of approx 7") appears to consist exclusively of dust. Both shells appear to have an over-all spherical geometry. The gas shell mass is estimated to be 2x10^-3 M(Sun), while the mass of the dust shell is estimated to be 5x10^-5 M(Sun). The derived expansion velocity, from the KI and NaD lines, of the gas shell, 19.5 km/s, agrees with that obtained from CO radio line data. The inferred shell age is 2700 years. There is structure, e.g. in the form of arcs, inside the gas shell, but it is not clear whether these are due to additional shells. Conclusions: Our results support the hypothesis that the observed geometrically thin, detached shells around carbon stars are the results of brief periods of intense mass loss, probably associated with thermal pulses, and subsequent wind-wind interactions.
|
|
| 31. |
- Maercker, Matthias, 1979, et al.
(författare)
-
The detached dust and gas shells around the carbon star U Antliae
- 2010
-
Ingår i: Astronomy and Astrophysics. - : EDP Sciences. - 0004-6361 .- 1432-0746. ; 511:1, s. A37-
-
Tidskriftsartikel (refereegranskat)abstract
- Context. Geometrically thin, detached shells of gas have been found around a handful of carbon stars. The current knowledge on these shells is mostly based on CO radio line data. However, imaging in scattered stellar light adds important new information as well as allows studies of the dust shells. Aims. Previous observations of scattered stellar light in the circumstellar medium around the carbon star U Ant were taken through filters centred on the resonance lines of K and Na. These observations could not separate the scattering by dust and atoms. The aim of this paper is to remedy this situation. Methods. We have obtained polarization data on stellar light scattered in the circumstellar medium around U Ant through filters which contain no strong lines, making it possible to differentiate between the two scattering agents. Kinematic, as well as spatial, information on the gas shells were obtained through high-resolution echelle spectrograph observations of the KI and NaD lines. Results. We confirm the existence of two detached shells around U Ant. The inner shell (at a radius of approximate to 43 '' and a width of approximate to 2 '') consists mainly of gas, while the outer shell (at a radius of approximate to 50 '' and a width of approximate to 7 '') appears to consist exclusively of dust. Both shells appear to have an over-all spherical geometry. The gas shell mass is estimated to be 2 x 10(-3) M-circle dot, while the mass of the dust shell is estimated to be 5 x 10(-5) M-circle dot. The derived expansion velocity, from the KI and NaD lines, of the gas shell, 19.5 km s(-1), agrees with that obtained from CO radio line data. The inferred shell age is 2700 years. There is structure, e.g. in the form of arcs, inside the gas shell, but it is not clear whether these are due to additional shells. Conclusions. Our results support the hypothesis that the observed geometrically thin, detached shells around carbon stars are the results of brief periods of intense mass loss, probably associated with thermal pulses, and subsequent wind-wind interactions. The separation into a gas and a dust shell, with different widths, is most likely the effect of different dynamical evolutions of the two media after their ejection.
|
|
| 32. |
- Mahajan, Anubha, et al.
(författare)
-
Refining the accuracy of validated target identification through coding variant fine-mapping in type 2 diabetes
- 2018
-
Ingår i: Nature Genetics. - : Nature Publishing Group. - 1061-4036 .- 1546-1718. ; 50:4, s. 559-571
-
Tidskriftsartikel (refereegranskat)abstract
- We aggregated coding variant data for 81,412 type 2 diabetes cases and 370,832 controls of diverse ancestry, identifying 40 coding variant association signals (P < 2.2 × 10−7); of these, 16 map outside known risk-associated loci. We make two important observations. First, only five of these signals are driven by low-frequency variants: even for these, effect sizes are modest (odds ratio ≤1.29). Second, when we used large-scale genome-wide association data to fine-map the associated variants in their regional context, accounting for the global enrichment of complex trait associations in coding sequence, compelling evidence for coding variant causality was obtained for only 16 signals. At 13 others, the associated coding variants clearly represent ‘false leads’ with potential to generate erroneous mechanistic inference. Coding variant associations offer a direct route to biological insight for complex diseases and identification of validated therapeutic targets; however, appropriate mechanistic inference requires careful specification of their causal contribution to disease predisposition.
|
|
| 33. |
- Olofsson, Hans, 1952, et al.
(författare)
-
High-resolution HST/ACS images of detached shells around carbon stars
- 2010
-
Ingår i: Astronomy and Astrophysics. - : EDP Sciences. - 0004-6361 .- 1432-0746. ; 515:JUN 2010, s. A270-
-
Tidskriftsartikel (refereegranskat)abstract
- Context. Overall spherically symmetric, geometrically thin gas and dust shells have been found around a handful of asymptotic giant branch (AGB) carbon stars. Their dynamical ages lie in the range of 10(3) to 10(4) years. A tentative explanation for their existence is that they have formed as a consequence of mass-loss-rate modulations during a He-shell flash. Aims. The detached shells carry information on their formation process, as well as on the small-scale structure of the circumstellar medium around AGB stars due to the absence of significant line-of-sight confusion. Methods. The youngest detached shells, those around the carbon stars R Scl and U Cam, are studied here in great detail in scattered stellar light with the Advanced Survey Camera on the Hubble Space Telescope. Quantitative results are derived assuming optically thin dust scattering. Results. The detached dust shells around R Scl and U Cam are found to be consistent with an overall spherical symmetry. They have radii of 19 ''.2 (corresponding to a linear size of 8 x 10(16) cm) and 7 ''.7 (5 x 10(16) cm), widths of 1 ''.2 (5 x 10(15) cm) and 0 ''.6 (4 x 10(15) cm), and dust masses of 3 x 10(-6) and 3 x 10(-7) M-circle dot, respectively. The dynamical ages of the R Scl and U Cam shells are estimated to be 1700 and 700 yr, respectively, and the shell widths correspond to time scales of 100 and 50 yr, respectively. Small-scale structure in the form of less than arcsec-sized clumps is clearly seen in the images of the R Scl shell. Average clump dust masses are estimated to be about 2 x 10(-9) M-circle dot. Comparisons with CO line interferometer data show that the dust and gas shells coincide spatially, within the errors (
|
|
| 34. |
- Sterky, Fredrik, et al.
(författare)
-
A Populus EST resource for plant functional genomics
- 2004
-
Ingår i: Proceedings of the National Academy of Sciences of the United States of America. - : Proceedings of the National Academy of Sciences. - 0027-8424 .- 1091-6490. ; 101:38, s. 13951-13956
-
Tidskriftsartikel (refereegranskat)abstract
- Trees present a life form of paramount importance for terrestrial ecosystems and human societies because of their ecological structure and physiological function and provision of energy and industrial materials. The genus Populus is the internationally accepted model for molecular tree biology. We have analyzed 102,019 Populus ESTs that clustered into 11,885 clusters and 12,759 singletons. We also provide >4,000 assembled full clone sequences to serve as a basis for the upcoming annotation of the Populus genome sequence. A public web-based EST database (POPULUSDB) provides digital expression profiles for 18 tissues that comprise the majority of differentiated organs. The coding content of Populus and Arabidopsis genomes shows very high similarity, indicating that differences between these annual and perennial angiosperm life forms result primarily from differences in gene regulation. The high similarity between Populus and Arabidopsis will allow studies of Populus to directly benefit from the detailed functional genomic information generated for Arabidopsis, enabling detailed insights into tree development and adaptation. These data will also valuable for functional genomic efforts in Arabidopsis.
|
|
| 35. |
- Vasluianu, Florin -Alexandru, et al.
(författare)
-
NTIRE 2023 Image Shadow Removal Challenge Report
- 2023
-
Ingår i: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). - Vancover : Institute of Electrical and Electronics Engineers (IEEE). ; , s. 1788-1807
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- This work reviews the results of the NTIRE 2023 Challenge on Image Shadow Removal. The described set of solutions were proposed for a novel dataset, which captures a wide range of object-light interactions. It consists of 1200 roughly pixel aligned pairs of real shadow free and shadow affected images, captured in a controlled environment. The data was captured in a white-box setup, using professional equipment for lights and data acquisition. The challenge had a number of 144 participants registered, out of which 19 teams were compared in the final ranking. The proposed solutions extend the work on shadow removal, improving over well-established state-of-the-art methods.
|
|
| 36. |
- Wang, Longguang, et al.
(författare)
-
NTIRE 2023 Challenge on Stereo Image Super-Resolution : Methods and Results
- 2023
-
Ingår i: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). - Vancover : Institute of Electrical and Electronics Engineers (IEEE). - 9798350302493 - 9798350302509 ; , s. 1346-1372
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- In this paper, we summarize the 2nd NTIRE challenge on stereo image super-resolution (SR) with a focus on new solutions and results. The task of the challenge is to super-resolve a low-resolution stereo image pair to a high-resolution one with a magnification factor of x4. Compared with single image SR, the major challenge of this challenge lies in how to exploit additional information in another viewpoint and how to maintain stereo consistency in the results. This challenge has 3 tracks, including one track on distortion (e.g., PSNR) and bicubic degradation, one track on perceptual quality (e.g., LPIPS) and bicubic degradation, as well as another track on real degradations. In total, 175, 93, and 103 participants were successfully registered for each track, respectively. In the test phase, 21, 17, and 12 teams successfully submitted results with PSNR (RGB) scores better than the baseline. This challenge establishes a new benchmark for stereo image SR.
|
|
