Poor O 2 supply to the infiltrated immune cells in the joint synovium of rheumatoid arthritis (RA) up-regulates hypoxia-inducible factor (HIF-1α) expression and induces reactive oxygen species (ROS) generation, both of which exacerbate synovial inflammation. Synovial inflammation in RA can be resolved by eliminating pro-inflammatory M1 macrophages and inducing anti-inflammatory M2 macrophages. Because hypoxia and ROS in the RA synovium play a crucial role in the induction of M1 macrophages and reduction of M2 macrophages, herein, we develop manganese ferrite and ceria nanoparticle-anchored mesoporous silica nanoparticles (MFC-MSNs) that can synergistically scavenge ROS and produce O 2 for reducing M1 macrophage levels and inducing M2 macrophages for RA treatment. MFC-MSNs exhibit a synergistic effect on O 2 generation and ROS scavenging that is attributed to the complementary reaction of ceria nanoparticles (NPs) that can scavenge intermediate hydroxyl radicals generated by manganese ferrite NPs in the process of O 2 generation during the Fenton reaction, leading to the efficient polarization of M1 to M2 macrophages both in vitro and in vivo. Intra-articular administration of MFC-MSNs to rat RA models alleviated hypoxia, inflammation, and pathological features in the joint. Furthermore, MSNs were used as a drug-delivery vehicle, releasing the anti-rheumatic drug methotrexate in a sustained manner to augment the therapeutic effect of MFC-MSNs. This study highlights the therapeutic potential of MFC-MSNs that simultaneously generate O 2 and scavenge ROS, subsequently driving inflammatory macrophages to the antiinflammatory subtype for RA treatment.

In object detection, determining which anchors to assign as positive or negative samples, known as anchor assignment, has been revealed as a core procedure that can significantly affect a model's performance. In this paper we propose a novel anchor assignment strategy that adaptively separates anchors into positive and negative samples for a ground truth bounding box according to the model's learning status such that it is able to reason about the separation in a probabilistic manner. To do so we first calculate the scores of anchors conditioned on the model and fit a probability distribution to these scores. The model is then trained with anchors separated into positive and negative samples according to their probabilities. Moreover, we investigate the gap between the training and testing objectives and propose to predict the Intersection-over-Unions of detected boxes as a measure of localization quality to reduce the discrepancy. The combined score of classification and localization qualities serving as a box selection metric in non-maximum suppression well aligns with the proposed anchor assignment strategy and leads significant performance improvements. The proposed methods only add a single convolutional layer to RetinaNet baseline and does not require multiple anchors per location, so are efficient. Experimental results verify the effectiveness of the proposed methods. Especially, our models set new records for single-stage detectors on MS COCO test-dev dataset with various backbones. Code is available at https://github.com/kkhoot/PAA.

No abstract

A supercooled liquid generally exhibits marked shear-thinning behavior, but its detailed mechanism remains elusive. Here we study the dynamics of structural rearrangements in supercooled liquids under shear, using two-dimensional (2D) molecular dynamics simulation. To elucidate the relationship between heterogeneous dynamics and the rheological behavior, we extend the four-point correlation function, which has been used for analyzing "dynamic heterogenity" in a quiescent condition, to a system under steady shear. In the Newtonian regime, the rearrangement dynamics is strongly heterogeneous in space, but remains isotropic. Contrary to this, in the non-Newtonian regime, where marked shear-thinning behavior appears, we find a novel dynamic effect: The mobile region tends to form anisotropic "fluidized bands." This finding suggests a link between nonlinear rheology and inhomogeneization of flow.

BACKGROUND Intestinal fibrosis causes many complications of Crohn’s disease (CD). Available biomarkers and imaging modalities lack sufficient accuracy to distinguish intestinal inflammation from fibrosis. Transcutaneous ultrasound elasticity imaging (UEI) is a promising, noninvasive approach for measuring tissue mechanical properties. We hypothesized that UEI could differentiate inflammatory from fibrotic bowel wall changes in both animal models of colitis and humans with CD. METHODS Female Lewis rats underwent weekly trinitrobenzene sulfonic acid enemas yielding models of acute inflammatory colitis (n = 5) and chronic intestinal fibrosis (n = 6). UEI scanning used a novel speckle-tracking algorithm to estimate tissue strain. Resected bowel segments were evaluated for evidence of inflammation and fibrosis. Seven consecutive patients with stenotic CD were studied with UEI and their resected stenotic and normal bowel segments were evaluated by ex vivo elastometry and histopathology. RESULTS Transcutaneous UEI normalized strain was able to differentiate acutely inflamed (−2.07) versus chronic fibrotic (−1.10) colon in rat models of inflammatory bowel disease (IBD; P = .037). Transcutaneous UEI normalized strain also differentiated stenotic (−0.87) versus adjacent normal small bowel (−1.99) in human CD (P = .0008), and this measurement also correlated well with ex vivo elastometry (r = −0.81). CONCLUSIONS UEI can differentiate inflammatory from fibrotic intestine in rat models of IBD and can differentiate between fibrotic and unaffected intestine in a pilot study in humans with CD. UEI represents a novel technology with potential to become a new objective measure of progression of intestinal fibrosis. Prospective clinical studies in CD are needed.