Dual Stream Long Short-Term Memory Feature Fusion Classifier for Surface Electromyography Gesture Recognition.

Gesture recognition using electromyography (EMG) signals has prevailed recently in the field of human-computer interactions for controlling intelligent prosthetics. Currently, machine learning and deep learning are the two most commonly employed methods for classifying hand gestures. Despite traditional machine learning methods already achieving impressive performance, it is still a huge amount of work to carry out feature extraction manually. The existing deep learning methods utilize complex neural network architectures to achieve higher accuracy, which will suffer from overfitting, insufficient adaptability, and low recognition accuracy. To improve the existing phenomenon, a novel lightweight model named dual stream LSTM feature fusion classifier is proposed based on the concatenation of five time-domain features of EMG signals and raw data, which are both processed with one-dimensional convolutional neural networks and LSTM layers to carry out the classification. The proposed method can effectively capture global features of EMG signals using a simple architecture, which means less computational cost. An experiment is conducted on a public DB1 dataset with 52 gestures, and each of the 27 subjects repeats every gesture 10 times. The accuracy rate achieved by the model is 89.66%, which is comparable to that achieved by more complex deep learning neural networks, and the inference time for each gesture is 87.6 ms, which can also be implied in a real-time control system. The proposed model is validated using a subject-wise experiment on 10 out of the 40 subjects in the DB2 dataset, achieving a mean accuracy of 91.74%. This is illustrated by its ability to fuse time-domain features and raw data to extract more effective information from the sEMG signal and select an appropriate, efficient, lightweight network to enhance the recognition results.

First report of Aloe root and stem rot caused by Phytophthora palmivora in Yunnan Province, China.

Aloe genus plants are perennial evergreen herb belonging to Liliaceae family which is widely used in food, medicine, beauty, and health care (Kumar et al. 2019). In August 2021, symptoms of root and stem rot was observed in approximately 20% of Aloe vera plantings in Yuanjiang County, Yunnan Province, China (23° 64' 53" N, 101° 99' 84" E). The most typical symptoms were stem and root rot, browning and necrosis of vascular tissues, gradual greening, and reddish-browning of leaves from bottom to top, abscission, and eventual plant death (Fig. S1). Therefore, to isolate and identify the pathogen, the plants showing the above symptoms were collected. The plant tissues were cut from the edges of root and stem lesions, followed by disinfection with 75% ethanol for 1 min, rinsed three times with sterilized distilled water, and cut into 3 × 3 mm small squares after excision of marginal tissues. The tissues were transferred to the oomycetes selective medium (Liu et al. 2022) and incubated at 28 °C in the dark for 3~5 days, and suspected colonies were purified. The colonies were then inoculated onto potato dextrose agar (PDA), V8-juice agar (V8), and oatmeal agar (OA) medium plates for morphological characteristics. Finally, 18 isolates with the same colonial and morphological characteristics were obtained from 30 lesioned tissue and one of them was named as ARP1. On PDA, V8 and OA medium plates, the ARP1 colonies were white. On PDA plate, the mycelia were dense and the colonies were petal-like; on V8 plate, the mycelia were cashmere and the colonies were radial or star-like. Whereas, on OA plate, the mycelia were cotton-like and the colonies were fluffy and radial (Fig. S2 A~C). Mycelium did not have septum with high branching and swelling. Sporangia were abundant, semi-papillate, varying in shape from ovoid-ellipsoid to long-ellipsoid, 18-26 × 45-63 µm (average: 22 × 54 µm, n = 30), sporangia released numerous zoospores from the papillate after maturation. The chlamydospores were spherical, 20-35 µm in diameter (average: 27.5 µm, n = 30) (Fig. S2 D~F). These morphological features were like those of the pathogenic species of the oomycetes (Chen et al. 2022). For the molecular characterization, the genomic DNA of the isolate was extracted using the cetyl trimethyl ammonium bromide method, and the translation elongation factor 1α (tef-1α) (Stielow et al. 2015), ß-tubulin (ß-tub) (Kroon et al. 2004) and internal transcribed spacer (ITS) (White et al. 1990) of isolated strain ARP1 were amplified using primer pairs EF1-1018F/EF1-1620R, TUBUF2/TUBUR1 and ITS1/ITS4, respectively. The tef-1α, ß-tub genes and ITS region of ARP1 were directly sequenced and their sequence information was deposited in GenBank under accession numbers OQ506129, OQ506127 and OQ449628. ARP1 was clustered on the same evolutionary branch with Phytophthora palmivora (Fig. S3). To confirm the pathogenicity of ARP1, the main root of A. vera was wounded to 1 cm long and 2 mm deep with a scalpel blade followed by inoculation with 50 ml suspension of ARP1 zoospores at a concentration of 1 × 106 spores / ml per potted plant, and an equal volume of water as control. All inoculated plants were placed in the greenhouse at 28°C, 12 h / 12 h light / dark. After 15 dpi, the inoculated plants showed typical symptoms of wilted and drooping leaves and stem and root rot, same as observed in the field condition (Fig. S4). After inoculation with ARP1, a strain with the same morphological and molecular characteristics as the original isolate was re-isolated, confirming Koch's postulates. To our knowledge, this is the first report of P. palmivora causing root and stem rot of A. vera in the study region. This disease could be a potential risk for aloe production and therefore appropriate management measures should be taken.

Cardiolipin-Dependent Mitophagy Guides Outcome after Traumatic Brain Injury.

Mitochondrial energy production is essential for normal brain function. Traumatic brain injury (TBI) increases brain energy demands, results in the activation of mitochondrial respiration, associated with enhanced generation of reactive oxygen species. This chain of events triggers neuronal apoptosis via oxidation of a mitochondria-specific phospholipid, cardiolipin (CL). One pathway through which cells can avoid apoptosis is via elimination of damaged mitochondria by mitophagy. Previously, we showed that externalization of CL to the mitochondrial surface acts as an elimination signal in cells. Whether CL-mediated mitophagy occurs in vivo or its significance in the disease processes are not known. In this study, we showed that TBI leads to increased mitophagy in the human brain, which was also detected using TBI models in male rats. Knockdown of CL synthase, responsible for de novo synthesis of CL, or phospholipid scramblase-3, responsible for CL translocation to the outer mitochondrial membrane, significantly decreased TBI-induced mitophagy. Inhibition of mitochondrial clearance by 3-methyladenine, mdivi-1, or phospholipid scramblase-3 knockdown after TBI led to a worse outcome, suggesting that mitophagy is beneficial. Together, our findings indicate that TBI-induced mitophagy is an endogenous neuroprotective process that is directed by CL, which marks damaged mitochondria for elimination, thereby limiting neuronal death and behavioral deficits.SIGNIFICANCE STATEMENT Traumatic brain injury (TBI) increases energy demands leading to activation of mitochondrial respiration associated with enhanced generation of reactive oxygen species and resultant damage to mitochondria. We demonstrate that the complete elimination of irreparably damaged organelles via mitophagy is activated as an early response to TBI. This response includes translocation of mitochondria phospholipid cardiolipin from the inner membrane to the outer membrane where externalized cardiolipin mediates targeted protein light chain 3-mediated autophagy of damaged mitochondria. Our data on targeting phospholipid scramblase and cardiolipin synthase in genetically manipulated cells and animals strongly support the essential role of cardiolipin externalization mechanisms in the endogenous reparative plasticity of injured brain cells. Furthermore, successful execution and completion of mitophagy is beneficial in the context of preservation of cognitive functions after TBI.

ERK1/2/mTOR/Stat3 pathway-mediated autophagy alleviates traumatic brain injury-induced acute lung injury.

Acute lung injury (ALI) is one of several complications in patients with traumatic brain injury (TBI). Autophagy is a primary homeostatic process that promotes cell survival under stress. Accumulating evidence implicates autophagy in the pathogenesis of ALI under various conditions. However, the role of autophagy in TBI-induced ALI remains unknown. The aim of this study was to adjust autophagy with pharmacological agents to determine its functional significance in TBI-induced ALI. Rats were preconditioned with autophagy promoter rapamycin or inhibitor 3-methyladenine before they were challenged with TBI. Extracellular signal-regulated kinase 1/2 (ERK1/2) inhibitor U0126, mechanistic target of rapamycin (mTOR) inhibitor rapamycin, and signal transducer and activator of transcription 3 (Stat3) inhibitor S31-201 were used to test the role of ERK1/2/mTOR/Stat3 signaling pathway in regulating autophagy. Autophagy is activated in lung tissues after TBI. Enhancement of autophagy suppressed apoptosis, inflammation and oxidative stress in lung tissues, which were activated after TBI, whereas inhibition of autophagy aggravated these critical pathological changes. Autophagy also improved TBI-induced impairment in pulmonary barrier function, oxygenation function and static compliance. Furthermore, TBI-induced autophagy was mediated by ERK1/2/mTOR/Stat3 pathway, which may serve to reduce ALI and improve pulmonary barrier function, oxygenation function and static compliance. These findings are important for the prevention and treatment of TBI-induced ALI.

Activation of bradykinin B2 receptor induced the inflammatory responses of cytosolic phospholipase A2 after the early traumatic brain injury.

Phospholipase A2 is a known aggravator of inflammation and deteriorates neurological outcomes after traumatic brain injury (TBI), however the exact inflammatory mechanisms remain unknown. This study investigated the role of bradykinin and its receptor, which are known initial mediators within inflammation activation, as well as the mechanisms of the cytosolic phospholipase A2 (cPLA2)-related inflammatory responses after TBI. We found that cPLA2 and bradykinin B2 receptor were upregulated after a TBI. Rats treated with the bradykinin B2 receptor inhibitor LF 16-0687 exhibited significantly less cPLA2 expression and related inflammatory responses in the brain cortex after sustaining a controlled cortical impact (CCI) injury. Both the cPLA2 inhibitor and the LF16-0687 improved CCI rat outcomes by decreasing neuron death and reducing brain edema. The following TBI model utilized both primary astrocytes and primary neurons in order to gain further understanding of the inflammation mechanisms of the B2 bradykinin receptor and the cPLA2 in the central nervous system. There was a stronger reaction from the astrocytes as well as a protective effect of LF16-0687 after the stretch injury and bradykinin treatment. The protein kinase C pathway was thought to be involved in the B2 bradykinin receptor as well as the cPLA2-related inflammatory responses. Rottlerin, a Protein Kinase C (PKC) δ inhibitor, decreased the activity of the cPLA2 activity post-injury, and LF16-0687 suppressed both the PKC pathway and the cPLA2 activity within the astrocytes. These results indicated that the bradykinin B2 receptor-mediated pathway is involved in the cPLA2-related inflammatory response from the PKC pathway.

Clinical Study on Complications of Intracranial Ruptured Aneurysm Embolization by Stent-Assisted Coil.

BACKGROUND The aim of this study was to retrospectively analyze the incidence of complications of intracranial complex aneurysms embolization by stent-assisted coils, and to investigate the causes of complications and corresponding treatment methods. MATERIAL AND METHODS A total of 71 patients with subarachnoid hemorrhage (SAH) underwent stent-assisted coil embolization from 2015 to 2018 were enrolled in this study. Among them, 59 cases were single aneurysm, 12 cases were multiple aneurysms (11 cases with 2 aneurysms and 1 case with 3 aneurysms), for a total of 84 aneurysms. All enrolled patients received stent angioplasty except for 1 case. RESULTS There were 62 aneurysms (73.81%) treated with complete tamponade, 21 aneurysms (25.00%) treated with near-total tamponade and 1 aneurysm (1.19%) treated with partial tamponade. All aneurysms were evaluated based on GOS (Glascow outcome scale): 55 cases had GOS of 5 scores, 12 cases had GOS of 4 scores, 3 cases had GOS of 3 scores, and 1 case had GOS of 1 score. There were 67 SAH patients with good prognosis (GOS of 4-5 scores). In our study, the incidence of complications was 12.7%. Three cases experienced acute thrombosis, 2 cases experienced aneurysm rupture during embolization, and 1 case experienced postoperative focal ischemic changes with mild neurological deficits. CONCLUSIONS Stent-assisted coil embolization is safe, effective, and feasible for the treatment of intracranial ruptured aneurysms. Patients had a favorable outcome of as high as 94.4%. However, clinical skills should be improved to reduce the occurrence of complications. Prompt and timely treatment for complications of intracranial ruptured aneurysm is also of great significance.

MicroRNA-98 Attenuates Cell Migration and Invasion in Glioma by Directly Targeting Pre-B Cell Leukemia Homeobox 3.

Glioblastoma multiforme (GBM) is the most common primary brain tumor in adults. The extraordinary invasion of human GBM into adjacent normal brain tissues contributes to treatment failure. However, the mechanisms that control this process remain poorly understood. Increasing evidence has demonstrated that microRNAs are strongly implicated in the migration and invasion of GBM. In this study, we found that microRNA-98 (miR-98) was markedly downregulated in human glioma tissues and cell lines. Functional experiments indicated that restored expression of miR-98 attenuated glioma cell invasion and migration, whereas depletion of miR-98 promoted glioma cell invasion and migration. Subsequent investigation showed that pre-B-cell leukemia homeobox 3 (PBX3), an important transcription factor that controls tumor invasion, was a direct and functional target of miR-98 in GBM cells. Consistently, an orthotopic mouse model also demonstrated the suppressive effects of miR-98 overexpression on tumor invasion and PBX3 expression. Silencing of PBX3 using small interfering RNA inhibited the migratory and invasive capacities of glioma cells, whereas reintroduction of PBX3 into glioma cells reversed the anti-invasive function of miR-98. Furthermore, depletion of PBX3 phenocopied the effects of miR-98 overexpression in vivo. Finally, quantitative real-time polymerase chain reaction results showed that miR-98 was negatively correlated with PBX3 expression in 24 glioma tissues. Thus, we propose that PBX3 modulation by miR-98 has an important role in regulating GBM invasion and may serve as therapeutic target for GBM.

Melatonin attenuates traumatic brain injury-induced inflammation: a possible role for mitophagy.

Melatonin functions as a crucial mediator of sterile neuroinflammation; however, the underlying mechanisms remain poorly understood. Dysfunctional mitochondria, a main source of reactive oxygen species, are impacted in inflammation activation. This study aimed to examine the effect of melatonin on inflammation via elimination of damaged mitochondria after controlled cortical impact, an in vivo model of traumatic brain injury (TBI). Here, we demonstrated that inhibition of mitophagy, the selective degradation of damaged mitochondria by autophagy, markedly enhanced inflammation induced by TBI. Melatonin treatment activated mitophagy through the mTOR pathway, then to attenuate TBI-induced inflammation. Furthermore, treatment with melatonin significantly ameliorated neuronal death and behavioral deficits after TBI, while 3-methyladenine reversed this effect by inhibiting mitophagy. Taken together, these results highlight a role for melatonin in protecting against TBI-triggered immunopathology, which is accomplished by negatively regulating inflammation activation and IL-1ß secretion via the autophagy of damaged mitochondria.

Efficient and Robust Point Cloud Registration via Heuristics-guided Parameter Search.

Estimating the rigid transformation with 6 degrees of freedom based on a putative 3D correspondence set is a crucial procedure in point cloud registration. Existing correspondence identification methods usually lead to large outlier ratios (> 95% is common), underscoring the significance of robust registration methods. Many researchers turn to parameter search-based strategies (e.g., Branch-and-Bround) for robust registration. Although related methods show high robustness, their efficiency is limited to the high-dimensional search space. This paper proposes a heuristics-guided parameter search strategy to accelerate the search while maintaining high robustness. We first sample some correspondences (i.e., heuristics) and then just need to sequentially search the feasible regions that make each sample an inlier. Our strategy largely reduces the search space and can guarantee accuracy with only a few inlier samples, therefore enjoying an excellent trade-off between efficiency and robustness. Since directly parameterizing the 6-dimensional nonlinear feasible region for efficient search is intractable, we construct a three-stage decomposition pipeline to reparameterize the feasible region, resulting in three lower-dimensional sub-problems that are easily solvable via our strategy. Besides reducing the searching dimension, our decomposition enables the leverage of 1-dimensional interval stabbing at all three stages for searching acceleration. Moreover, we propose a valid sampling strategy to guarantee our sampling effectiveness, and a compatibility verification setup to further accelerate our search. Extensive experiments on both simulated and real-world datasets demonstrate that our approach exhibits comparable robustness with state-of-the-art methods while achieving a significant efficiency boost.

Low-frequency magnetic field therapy for glioblastoma: Current advances, mechanisms, challenges and future perspectives.

BACKGROUND: Glioblastoma (GBM) is the most common malignant tumour of the central nervous system. Despite recent advances in multimodal GBM therapy incorporating surgery, radiotherapy, systemic therapy (chemotherapy, targeted therapy), and supportive care, the overall survival (OS) remains poor, and long-term survival is rare. Currently, the primary obstacles hindering the effectiveness of GBM treatment are still the blood-brain barrier and tumor heterogeneity. In light of its substantial advantages over conventional therapies, such as strong penetrative ability and minimal side effects, low-frequency magnetic fields (LF-MFs) therapy has gradually caught the attention of scientists. AIM OF REVIEW: In this review, we shed the light on the current status of applying LF-MFs in the treatment of GBM. We specifically emphasize our current understanding of the mechanisms by which LF-MFs mediate anticancer effects and the challenges faced by LF-MFs in treating GBM cells. Furthermore, we discuss the prospective applications of magnetic field therapy in the future treatment of GBM. Key scientific concepts of review: The review explores the current progress on the use of LF-MFs in the treatment of GBM with a special focus on the potential underlying mechanisms of LF-MFs in anticancer effects. Additionally, we also discussed the complex magnetic field features and biological characteristics related to magnetic bioeffects. Finally, we proposed a promising magnetic field treatment strategy for future applications in GBM therapy.

IDH1-mutant metabolite D-2-hydroxyglutarate inhibits proliferation and sensitizes glioma to temozolomide via down-regulating ITGB4/PI3K/AKT.

The heterogeneous molecular subtypes of gliomas demonstrate varied responses to chemotherapy and distinct prognostic outcomes. Gliomas with Isocitrate dehydrogenase 1 (IDH1) mutation are associated with better outcomes and are more responsive to temozolomide (TMZ) compared to those without IDH1 mutation. IDH1-mutant gliomas elevate D-2-hydroxyglutarate (D-2HG) levels, with potential dual effects on tumor progression. Limited research has explored the potential anti-glioma effects of D-2HG in combination with TMZ. Clinical data from over 2500 glioma patients in our study confirms that those with IDH1 mutations exhibit enhanced responsiveness to TMZ chemotherapy and a significantly better prognosis compared to IDH1 wild-type patients. In subsequent cellular experiments, we found that the IDH1-mutant metabolite D-2HG suppresses Integrin subunit beta 4 (ITGB4) expression, and down-regulate the phosphorylation levels of PI3K and AKT, ultimately inhibiting cell proliferation while promoting apoptosis, thereby improving glioma prognosis. Additionally, we have demonstrated the synergistic effect of D-2HG and TMZ in anti-glioma therapy involved inhibiting the proliferation of glioma cells and promoting apoptosis. Finally, by integrating data from the CGGA and TCGA databases, it was validated that ITGB4 expression was lower in IDH1-mutant gliomas, and patients with lower ITGB4 expression were associated with better prognosis. These findings indicate that ITGB4 may be a promising therapeutic target for gliomas and D-2HG inhibits proliferation and sensitizes glioma to temozolomide via down-regulating ITGB4/PI3K/AKT. These findings drive theoretical innovation and research progress in glioma therapy.

HRegNet: A Hierarchical Network for Efficient and Accurate Outdoor LiDAR Point Cloud Registration.

Point cloud registration is a fundamental problem in 3D computer vision. Outdoor LiDAR point clouds are typically large-scale and complexly distributed, which makes the registration challenging. In this paper, we propose an efficient hierarchical network named HRegNet for large-scale outdoor LiDAR point cloud registration. Instead of using all points in the point clouds, HRegNet performs registration on hierarchically extracted keypoints and descriptors. The overall framework combines the reliable features in deeper layer and the precise position information in shallower layers to achieve robust and precise registration. We present a correspondence network to generate correct and accurate keypoints correspondences. Moreover, bilateral consensus and neighborhood consensus are introduced for keypoints matching, and novel similarity features are designed to incorporate them into the correspondence network, which significantly improves the registration performance. In addition, we design a consistency propagation strategy to effectively incorporate spatial consistency into the registration pipeline. The whole network is also highly efficient since only a small number of keypoints are used for registration. Extensive experiments are conducted on three large-scale outdoor LiDAR point cloud datasets to demonstrate the high accuracy and efficiency of the proposed HRegNet. The source code of the proposed HRegNet is available at https://github.com/ispc-lab/HRegNet2.

Sparse-to-Dense Matching Network for Large-Scale LiDAR Point Cloud Registration.

Point cloud registration is a fundamental problem in 3D computer vision. Previous learning-based methods for LiDAR point cloud registration can be categorized into two schemes: dense-to-dense matching methods and sparse-to-sparse matching methods. However, for large-scale outdoor LiDAR point clouds, solving dense point correspondences is time-consuming, whereas sparse keypoint matching easily suffers from keypoint detection error. In this paper, we propose SDMNet, a novel Sparse-to-Dense Matching Network for large-scale outdoor LiDAR point cloud registration. Specifically, SDMNet performs registration in two sequential stages: sparse matching stage and local-dense matching stage. In the sparse matching stage, we sample a set of sparse points from the source point cloud and then match them to the dense target point cloud using a spatial consistency enhanced soft matching network and a robust outlier rejection module. Furthermore, a novel neighborhood matching module is developed to incorporate local neighborhood consensus, significantly improving performance. The local-dense matching stage is followed for fine-grained performance, where dense correspondences are efficiently obtained by performing point matching in local spatial neighborhoods of high-confidence sparse correspondences. Extensive experiments on three large-scale outdoor LiDAR point cloud datasets demonstrate that the proposed SDMNet achieves state-of-the-art performance with high efficiency.

Ecological Risks of Antibiotics in Urban Wetlands on the Qinghai-Tibet Plateau, China.

Although the ecological risks of antibiotics have been extensively researched globally, fewer studies have been conducted in sensitive and fragile plateau wetland ecosystems. To evaluate the ecological risk of antibiotics in plateau urban wetlands, 18 water samples, 10 plant samples, and 8 sediment samples were collected in March 2022 in the Xining urban wetlands on the Qinghai-Tibet Plateau. The liquid chromatography-electrospray ionization tandem mass spectrometry method was utilized to measure the concentrations of 15 antibiotics in three categories in three types of environmental media. Risk quotients were adopted to assess the ecological risk of antibiotics, and the principal component analysis-multiple linear regression model was used to analyze the source of antibiotics. The results showed that (1) the maximum concentrations of antibiotics in water samples, plants, and sediments reached 1220.86 ng/L, 78.30 ng/g, and 5.64 ng/g, respectively; (2) Tylosin (TYL), norfloxacin (NFX), ofloxacin (OFX), and ciprofloxacin (CFX) in water were at medium and high-risk levels, and OFX had the highest risk value, of 108.04; and (3) the results of source apportionment indicate that 58.94% of the antibiotics came from the Huangshui river and wastewater treatment plant (WWTP) near the wetlands. The current study may provide a reference for the risks and management of antibiotics in plateau urban wetlands.

Globally Optimal Vertical Direction Estimation in Atlanta World.

In man-made environments, most of the objects and structures are organized in the form of orthogonal and parallel planes. These planes can be approximated by an Atlanta world assumption, in which the normals of planes can be represented by Atlanta frames. The Atlanta world assumption has one vertical frame and multiple horizontal frames. Conventionally, given a set of inputs such as surface normals, the Atlanta frame estimation problem can be solved by a branch-and-bound (BnB) algorithm. However, the runtime of the BnB algorithm will increase greatly when the dimensionality (i.e., the number of horizontal frames) increases. In this paper, we estimate only the vertical direction, instead of all Atlanta frames at once. Accordingly, we propose a vertical direction estimation method by considering the relationship between the vertical frame and horizontal frames. Concretely, our approach employs a BnB algorithm to search the vertical direction, thereby guaranteeing global optimality without requiring prior knowledge of the number of Atlanta frames. In order to guarantee convergence, four novel bounds are investigated, by mapping a 3D hemisphere to a 2D region. We verify the feasibility of the proposed method using various challenging synthetic and real-world data.

Efficient and Outlier-Robust Simultaneous Pose and Correspondence Determination by Branch-and-Bound and Transformation Decomposition.

Estimating the pose of a calibrated camera relative to a 3D point set from one image is an important task in computer vision. Perspective-n-Point algorithms are often used if perfect 2D-3D correspondences are known. However, it is difficult to determine 2D-3D correspondences perfectly, and then the simultaneous pose and correspondence determination problem is needed to be solved. Early methods aimed to solve this problem by local optimization. Recently, several new methods are proposed to globally solve this problem by using branch-and-bound (BnB) method, but they tend to be slow because the time complexity of the BnB-based methods is exponential to the dimensionality of the parameter space, and they directly search the 6D parameter space. In this paper, we propose to decompose the joint searching into two separate searching processes by introducing a rotation invariant feature (RIF). Specifically, we construct RIFs from the original 3D and 2D point sets and search for the globally optimal translation to match these two RIFs first. Then, the original 3D point set is translated and matched with the 2D point set to find a globally optimal rotation. Experiments on challenging data show that the proposed method outperforms state-of-the-art methods in terms of both speed and accuracy.

Globally-Optimal Inlier Maximization for Relative Pose Estimation Under Planar Motion.

Planar motion constraint occurs in visual odometry (VO) and SLAM for Automated Guided Vehicles (AGVs) or mobile robots in general. Conventionally, two-point solvers can be nested to RANdom SAmple Consensus to reject outliers in real data, but the performance descends when the ratio of outliers goes high. This study proposes a globally-optimal Branch-and-Bound (BnB) solver for relative pose estimation under general planar motion, which aims to figure out the globally-optimal solution even under a quite noisy environment. Through reasonable modification of the motion equation, we decouple the relative pose into relative rotation and translation so that a simplified bounding strategy can be applied. It enhances the efficiency of the BnB technique. Experimental results support the global optimality and demonstrate that the proposed method performs more robustly than existing approaches. In addition, the proposed algorithm outperforms state-of-art methods in global optimality under the varying level of outliers.

miR-720 is a key regulator of glioma migration and invasion by controlling TARSL2 expression.

Glioblastoma (GBM) is the most lethal type of primary brain tumor and is characterized by diffuse infiltrative growth. However, the mechanisms that control this phenotype remain largely unknown. Emerging evidence has demonstrated that the abnormal expression of microRNAs and their target genes are involved in the migration and invasion of glioma cells. In this study, we demonstrated that microRNA-720 (miR-720) was significantly upregulated in glioma tissues and cells. Functional experiments showed that overexpression of miR-720 promotes glioma migration and invasion, while downregulation of miR-720 inhibits glioma migration and invasion. Meanwhile, we found that threonyl-tRNA synthetase like-2 (TARSL2) was a direct and functional target of miR-720 in glioma. Reintroduction of TARSL2 into glioma cells repressed the invasion promoting function of miR-720, whereas downregulation of TARSL2 reversed the anti-invasion function of anti-miR-720. Furthermore, quantitative real-time polymerase chain reaction results showed that miR-720 was inversely correlated with TARSL2 expression in 40 GBM tissues. Finally, in vivo experiments showed that miR-720 promotes glioma growth and upregulates invasion-related genes in nude mice. Overall, our findings suggest increasing miR-720 enhances glioma migration and invasion through downregulation of TARSL2, which may provide novel insight into the treatment of glioma.

Prokineticin-2 prevents neuronal cell deaths in a model of traumatic brain injury.

Prokineticin-2 (Prok2) is an important secreted protein likely involved in the pathogenesis of several acute and chronic neurological diseases through currently unidentified regulatory mechanisms. The initial mechanical injury of neurons by traumatic brain injury triggers multiple secondary responses including various cell death programs. One of these is ferroptosis, which is associated with dysregulation of iron and thiols and culminates in fatal lipid peroxidation. Here, we explore the regulatory role of Prok2 in neuronal ferroptosis in vitro and in vivo. We show that Prok2 prevents neuronal cell death by suppressing the biosynthesis of lipid peroxidation substrates, arachidonic acid-phospholipids, via accelerated F-box only protein 10 (Fbxo10)-driven ubiquitination, degradation of long-chain-fatty-acid-CoA ligase 4 (Acsl4), and inhibition of lipid peroxidation. Mice injected with adeno-associated virus-Prok2 before controlled cortical impact injury show reduced neuronal degeneration and improved motor and cognitive functions, which could be inhibited by Fbxo10 knockdown. Our study shows that Prok2 mediates neuronal cell deaths in traumatic brain injury via ferroptosis.