Search

The development of generative models that create 3D content from a text prompt has made considerable strides thanks to the use of the score distillation sampling (SDS) method on pre-trained diffusion models for image generation. However, the SDS method is also the source of several artifacts, such as the Janus problem, the misalignment between the text prompt and the generated 3D model, and 3D model inaccuracies. While existing methods heavily rely on the qualitative assessment of these artifacts through visual inspection of a limited set of samples, in this work we propose more objective quantitative evaluation metrics, which we cross-validate via human ratings, and show analysis of the failure cases of the SDS technique. We demonstrate the effectiveness of this analysis by designing a novel computationally efficient baseline model that achieves state-of-the-art performance on the proposed metrics while addressing all the above-mentioned artifacts.

In this paper, we propose an effective two-stage approach named Grounded-Dreamer to generate 3D assets that can accurately follow complex, compositional text prompts while achieving high fidelity by using a pre-trained multi-view diffusion model. Multi-view diffusion models, such as MVDream, have shown to generate high-fidelity 3D assets using score distillation sampling (SDS). However, applied naively, these methods often fail to comprehend compositional text prompts, and may often entirely omit certain subjects or parts. To address this issue, we first advocate leveraging text-guided 4-view images as the bottleneck in the text-to-3D pipeline. We then introduce an attention refocusing mechanism to encourage text-aligned 4-view image generation, without the necessity to re-train the multi-view diffusion model or craft a high-quality compositional 3D dataset. We further propose a hybrid optimization strategy to encourage synergy between the SDS loss and the sparse RGB reference images. Our method consistently outperforms previous state-of-the-art (SOTA) methods in generating compositional 3D assets, excelling in both quality and accuracy, and enabling diverse 3D from the same text prompt., Comment: 9 pages, 10 figures

Single-cell technologies have transformed our understanding of human tissues. Yet, studies typically capture only a limited number of donors and disagree on cell type definitions. Integrating many single-cell datasets can address these limitations of individual studies and capture the variability present in the population. Here we present the integrated Human Lung Cell Atlas (HLCA), combining 49 datasets of the human respiratory system into a single atlas spanning over 2.4 million cells from 486 individuals. The HLCA presents a consensus cell type re-annotation with matching marker genes, including annotations of rare and previously undescribed cell types. Leveraging the number and diversity of individuals in the HLCA, we identify gene modules that are associated with demographic covariates such as age, sex and body mass index, as well as gene modules changing expression along the proximal-to-distal axis of the bronchial tree. Mapping new data to the HLCA enables rapid data annotation and interpretation. Using the HLCA as a reference for the study of disease, we identify shared cell states across multiple lung diseases, including SPP1+ profibrotic monocyte-derived macrophages in COVID-19, pulmonary fibrosis and lung carcinoma. Overall, the HLCA serves as an example for the development and use of large-scale, cross-dataset organ atlases within the Human Cell Atlas.

Single-cell technologies have transformed our understanding of human tissues. Yet, studies typically capture only a limited number of donors and disagree on cell type definitions. Integrating many single-cell datasets can address these limitations of individual studies and capture the variability present in the population. Here we present the integrated Human Lung Cell Atlas (HLCA), combining 49 datasets of the human respiratory system into a single atlas spanning over 2.4 million cells from 486 individuals. The HLCA presents a consensus cell type re-annotation with matching marker genes, including annotations of rare and previously undescribed cell types. Leveraging the number and diversity of individuals in the HLCA, we identify gene modules that are associated with demographic covariates such as age, sex and body mass index, as well as gene modules changing expression along the proximal-to-distal axis of the bronchial tree. Mapping new data to the HLCA enables rapid data annotation and interpretation. Using the HLCA as a reference for the study of disease, we identify shared cell states across multiple lung diseases, including SPP1+ profibrotic monocyte-derived macrophages in COVID-19, pulmonary fibrosis and lung carcinoma. Overall, the HLCA serves as an example for the development and use of large-scale, cross-dataset organ atlases within the Human Cell Atlas.

We describe a first step towards learning general-purpose visual representations of physical scenes using only image prediction as a training criterion. To do so, we first define "physical scene" and show that, even though different agents may maintain different representations of the same scene, the underlying physical scene that can be inferred is unique. Then, we show that NeRFs cannot represent the physical scene, as they lack extrapolation mechanisms. Those, however, could be provided by Diffusion Models, at least in theory. To test this hypothesis empirically, NeRFs can be combined with Diffusion Models, a process we refer to as NeRF Diffusion, used as unsupervised representations of the physical scene. Our analysis is limited to visual data, without external grounding mechanisms that can be provided by independent sensory modalities., Comment: TLDR: Physical scenes are equivalence classes of sufficient statistics, and can be inferred uniquely by any agent measuring the same finite data; We formalize and implement an approach to representation learning that overturns "naive realism" in favor of an analytical approach of Russell and Koenderink. NeRFs cannot capture the physical scenes, but combined with Diffusion Models they can

INTRODUCTION: Sports-related concussion (SRC) is a common form of brain injury that lacks reliable methods to guide clinical decisions. MicroRNAs (miRNAs) can influence biological processes involved in SRC, and measurement of miRNAs in biological fluids may provide objective diagnostic and return to play/recovery biomarkers. Therefore, this prospective study investigated the temporal profile of circulating miRNA levels in concussed male and female athletes. METHODS: Pre-season baseline blood samples were collected from amateur Australian rules football players (82 males, 45 females). Of these, 20 males and 8 females sustained an SRC during the subsequent season and underwent blood sampling at 2-, 6- and 13-days post-injury. A miRNA discovery Open Array was conducted on plasma to assess the expression of 754 known/validated miRNAs. miRNA target identified were further investigated with quantitative real-time PCR (qRT-PCR) in a validation study. Data pertaining to SRC symptoms, demographics, sporting history, education history and concussion history were also collected. RESULTS: Discovery analysis identified 18 candidate miRNA. The consequent validation study found that plasma miR-221-3p levels were decreased at 6d and 13d, and that miR-27a-3p levels were decreased at 6d, when compared to baseline. Moreover, miR-27a and miR-221-3p levels were inversely correlated with SRC symptom severity. CONCLUSION: Circulating levels of miR-27a-3p and miR-221-3p were decreased in the sub-acute stages after SRC, and were inversely correlated with SRC symptom severity. Although further studies are required, these analyses have identified miRNA biomarker candidates of SRC severity and recovery that may one day assist in its clinical management.

In this age of social media, people often look at what others are wearing. In particular, Instagram and Twitter influencers often provide images of themselves wearing different outfits and their followers are often inspired to buy similar clothes.We propose a system to automatically find the closest visually similar clothes in the online Catalog (street-to-shop searching). The problem is challenging since the original images are taken under different pose and lighting conditions. The system initially localizes high-level descriptive regions (top, bottom, wristwear. . . ) using multiple CNN detectors such as YOLO and SSD that are trained specifically for apparel domain. It then classifies these regions into more specific regions such as t-shirts, tunic or dresses. Finally, a feature embedding learned using a multi-task function is recovered for every item and then compared with corresponding items in the online Catalog database and ranked according to distance. We validate our approach component-wise using benchmark datasets and end-to-end using human evaluation., Comment: KDD2019, AI for Fashion Workshop

Vascularization is a major hurdle for growing three-dimensional tissue engineered constructs. This study investigated the mechanisms involved in hypoxic preconditioning of primary rat myoblasts in vitro and their influence on local angiogenesis postimplantation. Primary rat myoblast cultures were exposed to 90 min hypoxia at <1% oxygen followed by normoxia for 24 h. Real time (RT) polymerase chain reaction evaluation indicated that 90 min hypoxia resulted in significant downregulation of miR-1 and miR-206 (p < 0.05) and angiopoietin-1 (p < 0.05) with upregulation of vascular endothelial growth factor-A (VEGF-A; p < 0.05). The miR-1 and angiopoietin-1 responses remained significantly downregulated after a 24 h rest phase. In addition, direct inhibition of miR-206 in L6 myoblasts caused a significant increase in VEGF-A expression (p < 0.05), further establishing that changes in VEGF-A expression are influenced by miR-206. Of the myogenic genes examined, MyoD was significantly upregulated, only after 24 h rest (p < 0.05). Preconditioned or control myoblasts were implanted with Matrigel™ into isolated bilateral tissue engineering chambers incorporating a flow-through epigastric vascular pedicle in severe combined immunodeficiency mice and the chamber tissue harvested 14 days later. Chambers implanted with preconditioned myoblasts had a significantly increased percentage volume of blood vessels (p = 0.0325) compared with chambers implanted with control myoblasts. Hypoxic preconditioned myoblasts promote vascularization of constructs via VEGF upregulation and downregulation of angiopoietin-1, miR-1 and miR-206. The relatively simple strategy of hypoxic preconditioning of implanted cells - including non-stem cell types - has broad, future applications in tissue engineering of skeletal muscle and other tissues, as a technique to significantly increase implant site angiogenesis.

Natural progression of HIV-1 infection depends on genetic variation in the human major histocompatibility complex (MHC) class I locus, and the CD8+ T cell response is thought to be a primary mechanism of this effect. However, polymorphism within the MHC ma

Natural progression of HIV-1 infection depends on genetic variation in the human major histocompatibility complex (MHC) class I locus, and the CD8+ T cell response is thought to be a primary mechanism of this effect. However, polymorphism within the MHC ma

Natural progression of HIV-1 infection depends on genetic variation in the human major histocompatibility complex (MHC) class I locus, and the CD8+ T cell response is thought to be a primary mechanism of this effect. However, polymorphism within the MHC may also alter innate immune activity against human immunodeficiency virus type 1 (HIV-1) by changing interactions of human leukocyte antigen (HLA) class I molecules with leukocyte immunoglobulin-like receptors (LILR), a group of immunoregulatory receptors mainly expressed on myelomonocytic cells including dendritic cells (DCs). We used previously characterized HLA allotype-specific binding capacities of LILRB1 and LILRB2 as well as data from a large cohort of HIV-1-infected individuals (N = 5126) to test whether LILR-HLA class I interactions influence viral load in HIV-1 infection. Our analyses in persons of European descent, the largest ethnic group examined, show that the effect of HLA-B alleles on HIV-1 control correlates with the binding strength between corresponding HLA-B allotypes and LILRB2 (p = 10-2). Moreover, overall binding strength of LILRB2 to classical HLA class I allotypes, defined by the HLA-A/B/C genotypes in each patient, positively associates with viral replication in the absence of therapy in patients of both European (p = 10-11-10-9) and African (p = 10-5-10-3) descent. This effect appears to be driven by variations in LILRB2 binding affinities to HLA-B and is independent of individual class I allelic effects that are not related to the LILRB2 function. Correspondingly, in vitro experiments suggest that strong LILRB2-HLA binding negatively affects antigen-presenting properties of DCs. Thus, we propose an impact of LILRB2 on HIV-1 disease outcomes through altered regulation of DCs by LILRB2-HLA engagement.

Natural progression of HIV-1 infection depends on genetic variation in the human major histocompatibility complex (MHC) class I locus, and the CD8+ T cell response is thought to be a primary mechanism of this effect. However, polymorphism within the MHC may also alter innate immune activity against human immunodeficiency virus type 1 (HIV-1) by changing interactions of human leukocyte antigen (HLA) class I molecules with leukocyte immunoglobulin-like receptors (LILR), a group of immunoregulatory receptors mainly expressed on myelomonocytic cells including dendritic cells (DCs). We used previously characterized HLA allotype-specific binding capacities of LILRB1 and LILRB2 as well as data from a large cohort of HIV-1-infected individuals (N = 5126) to test whether LILR-HLA class I interactions influence viral load in HIV-1 infection. Our analyses in persons of European descent, the largest ethnic group examined, show that the effect of HLA-B alleles on HIV-1 control correlates with the binding strength between corresponding HLA-B allotypes and LILRB2 (p = 10-2). Moreover, overall binding strength of LILRB2 to classical HLA class I allotypes, defined by the HLA-A/B/C genotypes in each patient, positively associates with viral replication in the absence of therapy in patients of both European (p = 10-11-10-9) and African (p = 10-5-10-3) descent. This effect appears to be driven by variations in LILRB2 binding affinities to HLA-B and is independent of individual class I allelic effects that are not related to the LILRB2 function. Correspondingly, in vitro experiments suggest that strong LILRB2-HLA binding negatively affects antigen-presenting properties of DCs. Thus, we propose an impact of LILRB2 on HIV-1 disease outcomes through altered regulation of DCs by LILRB2-HLA engagement.

NADPH oxidases contribute to brain injury, yet they may also have a role in brain repair, particularly in vascular signaling and angiogenesis. This study determined the temporal and spatial profile of NADPH oxidase subunit expression/activity concurrently with angiogenesis in the brain following transient ischemic stroke induced by prolonged constriction of the middle cerebral artery by perivascular injection of endothelin-1 in conscious Hooded Wistar rats (n = 47). VEGF mRNA expression was increased in the ipsilateral cortex and striatum between 6 h and 28 days post-stroke concurrently with a marked increase in Nox2 mRNA expression up to 7 days, and increased Nox4 mRNA expression detected between 7 and 28 days. Point counting of blood vessels using Metamorph imaging software showed increased vascular sprouting between 3 and 7 days after stroke with new vascular networks detected in the core infarct region by 14 days. Angiogenic blood vessels 3 and 7 days post-stroke were observed to co-localise with both Nox2 antibody and dihydroethidium fluorescence suggesting a role for Nox2 generated superoxide during the phase of vascular remodeling, whilst Nox4 expression was detected once new cerebral vessels had formed. These results indicate for the first time that ROS signaling through a cerebrovascular Nox2 NADPH oxidase may be important in initiating brain angiogenesis.

MicroRNAs (miRNAs) are small noncoding RNAs that act as regulators of gene expression by targeting mature messenger RNAs. Following the initial report of the presence of miRNAs in serum and plasma a number of studies have successfully demonstrated the use of these miRNAs as biomarkers of disease. Currently, there are many methods of isolating total RNA from liquid samples. Here, we describe a simple, cost effective method for extraction of RNA from human serum as well as subsequent real time PCR analysis of miRNA levels.