LINC01305 recruits basonuclin 1 to act on G-protein pathway suppressor 1 to promote esophageal squamous cell carcinoma.

EsophageaL squamous cell carcinoma (ESCC) is one of the most common and lethal tumors, however, its underlying molecular mechanisms are not completely understood and new therapeutic targets are needed. Here, we found that the transcription factor basonuclin 1 (BNC1) was significantly upregulated and closely related to the differentiation and metastasis of ESCC. Furthermore, BNC1, LINC01305, and G-protein pathway suppressor 1 (GPS1) had significant oncogenic roles in ESCC. In addition, in vivo experiments showed that knockdown of BNC1 indeed significantly inhibited the proliferation and metastasis of ESCC. We also revealed the molecular mechanism by which LINC01305 recruits BNC1 to the promoter of GPS1, and then GPS1 could mediate the JNK signaling pathway to promote the proliferation and metastases of ESCC. Taken together, we discovered the novel molecular mechanism by which LINC01305/BNC1 upregulates GPS1 expression to promote the development of ESCC, providing a new therapeutic target for ESCC.

Effects of Plant-Derived Polyphenols on the Antioxidant Activity and Aroma of Sulfur-Dioxide-Free Red Wine.

Significant efforts have been made in recent years to produce healthier wines, with the primary goal of reducing the use of sulfur dioxide (SO2), which poses health risks. This study aimed to assess the effectiveness of three plant-derived polyphenols (dihydromyricetin, resveratrol, and catechins) as alternatives to SO2 in wine. After a three-month aging process, the wines were evaluated using analytical techniques such as high-performance liquid chromatography, colorimetry, gas chromatography-olfactometry-mass spectrometry, as well as electronic nose and electronic tongue analyses, with the purpose to assess parameters including antioxidant activity, color, contents of volatile aroma compounds, and sensory characteristics. The results demonstrated various degrees of improvement in the antioxidant activity, aromatic intensity, and sensory characteristics of wines using polyphenols. Notably, dihydromyricetin (200 mg/L) exhibited the strongest antioxidant activity, with increases of 18.84%, 23.28%, and 20.87% in 2,2-diphenyl-1-picrylhydrazyl, 2,2'azino-bis(3-ethylbenzothiazoline-6-sulfonic acid), and ferric-ion-reducing antioxidant power assays, respectively. Resveratrol (200 mg/L) made the most significant contribution to volatile aroma compounds, with an 8.89% increase in the total content of alcohol esters. In E-nose analysis, catechins (200 mg/L) showed the highest response to aromatic compounds and the lowest response to volatile sulfur compounds, while also exhibiting the best sensory characteristics. Therefore, the three plant-derived polyphenols investigated here exhibited the potential to enhance wine quality as alternatives to SO2. However, it is important to consider the specific impact of different polyphenols on wine; hence, suitable antioxidants should be selected in wine production according to specific requirements.

SCARA5 induced ferroptosis to effect ESCC proliferation and metastasis by combining with Ferritin light chain.

BACKGROUND: Esophageal squamous cell carcinoma (ESCC) remains one of the most lethal cancers worldwide accompany with an extremely poor prognosis. Therefore, this study aims to screen for new molecules affecting ESCC and explore their mechanisms of action to provide ideas for targeted therapies for ESCC. METHODS: Firstly, we screened out the membrane protein SCARA5 by high-throughput sequencing of the ESCC patient tissues, and RT-qPCR and WB were used to verify the differential expression of SCARA5 in esophageal cell lines, and IHC analyzed the expression localization of SCARA5 in ESCC tissue. Then, flow cytometry, wound healing assay, Transwell assay and CCK-8 assay were used to explore the effects of SCARA5 on cell cycle, migration and invasion as well as cell proliferation activity of esophageal squamous carcinoma cells. Meanwhile, transmission electron microscopy was used to detect changes in cellular mitochondrial morphology, and flow cytometry were used to detect changes in intracellular reactive oxygen metabolism, and immunofluorescence and flow cytometry were used to detect changes in intracellular Fe2+. Mechanistically, co-immunoprecipitation was used to detect whether SCARA5 binds to ferritin light chain, and ferroptosis-related protein expression was detected by WB. Finally, the tumor xenograft model was applied to validation the role of SCARA5 tumor growth inhibition in vivo. RESULTS: We found that SCARA5 was aberrantly decreased in ESCC tissues and cell lines. Furthermore, we confirmed that SCARA5 suppressed the cell cycle, metastasis and invasion of ESCC cells. Meanwhile, we also found that overexpression of SCARA5 caused changes in mitochondrial morphology, accumulation of intracellular reactive oxygen species and increased intracellular Fe2+ in ESCC cells, which induced ferroptosis in ESCC cells. Mechanically, we validated that SCARA5 combined with ferritin light chain and increased intracellular Fe2+. As well as, overexpression SCARA5 induced ferroptosis by increasing ferritin light chain in nude mice subcutaneous tumors and inhibited the growth of nude mice subcutaneous tumors. CONCLUSION: Collectively, our findings demonstrated that SCARA5 suppressed the proliferation and metastasis of ESCC by triggering ferroptosis through combining with ferritin light chain.

Degeneration-Aware Localization with Arbitrary Global-Local Sensor Fusion.

Global localization is a fundamental ability for mobile robots. Considering the limitation of single type of sensor, fusing measurements from multiple sensors with complementary properties is a valuable task for study. In this paper, we propose a decoupled optimization-based framework for global-local sensor fusion, which fuses the intermittent 3D global positions and high-frequent 6D odometry poses to infer the 6D global localization results in real-time. The fusion process is formulated as estimating the relative transformation between global and local reference coordinates, translational extrinsic calibration, and the scale of the local pose estimator. We validate the full observability of the system under general movements, and further analyze the degenerated movement patterns where some related system state would be unobservable. A degeneration-aware sensor fusion method is designed which detects the degenerated directions before optimization, and adds constraints specifically along these directions to relieve the effect of the noise. The proposed degeneration-aware global-local sensor fusion method is validated in both simulation and real-world datasets with different sensor configurations, and shows its effectiveness in terms of accuracy and robustness compared with other decoupled sensor fusion methods for global localization.

Systematic profiling identifies PDLIM2 as a novel prognostic predictor for oesophageal squamous cell carcinoma (ESCC).

Till now, no appropriate biomarkers for high-risk population screening and prognosis prediction have been identified for patients with oesophageal squamous cell carcinoma (ESCC). In this study, by the combined use of data from the Gene Expression Omnibus (GEO) datasets and The Cancer Genome Atlas (TCGA)-oesophageal carcinoma (ESCA), we aimed to screen dysregulated genes with prognostic value in ESCC and the genetic and epigenetic alterations underlying the dysregulation. About 222 genes that had at least fourfold change in ESCC compared with adjacent normal tissues were identified using the microarray data in GDS3838. Among these genes, only PDLIM2 was associated with nodal invasion and overall survival (OS) at the same time. The high PDLIM2 expression group had significantly longer OS and its expression was independently associated with better OS (HR: 0.64, 95% CI: 0.43-0.95, P = 0.03), after adjustment for gender and pathologic stages. The expression of its exon 7/8/9/10 had the highest AUC value (0.724) and better prognostic value (HR: 0.43, 95% CI: 0.22-0.83, P = 0.01) than total PDLIM2 expression. PDLIM2 DNA copy deletion was common in ESCC and was associated with decreased gene expression. The methylation status of two CpG sites (cg23696886 and cg20449614) in the proximal promoter region of PDLIM2 showed a moderate negative correlation with the gene expression in PDLIM2 copy neutral/amplification group. In conclusion, we infer that PDLIM2 expression might be a novel prognostic indicator for ESCC patients. Its exon 7/8/9/10 expression had the best prognostic value. Its down-regulation might be associated with gene-level copy deletion and promoter hypermethylation.

ENST00000489707.5 Is a Preferred Alternative Splicing Variant of PTK7 in Adrenocortical Cancer and Shows Potential Prognostic Value.

BACKGROUND This study aimed to explore the transcript preference of PTK7 in adrenocortical cancer (ACC), the prognostic value, and the potential underlying genetic alterations. MATERIAL AND METHODS Data from the Cancer Genome Atlas-Adrenocortical Cancer (TCGA-ACC) and the Genotype-Tissue Expression (GTEx)-normal adrenal gland were used for analysis. RESULTS A non-canonical alternative transcript, ENST00000489707.5, which only encodes an extracellular immunoglobulin (Ig)-like domain and an intracellular kinase domain, is the dominant isoform of PTK7 in both ACC and normal adrenal gland. Its expression percentage was significantly higher in ACC than in normal adrenal gland. ACC tissues showed preferred expression of this transcript compared with other cancers with known PTK7 expression. Prognostic analysis showed that ENST00000489707.5 had independent prognostic value in progression-free survival (PFS) (HR: 1.227, 95%CI: 1.077-1.398, p=0.002) and disease-specific survival (DSS) (HR: 1.419, 95%CI: 1.154-1.745, p=0.001) after adjustment of other risk factors. cg20819617 methylation was negatively correlated with both PTK7 and ENST00000489707.5 expression. CONCLUSIONS ENST00000489707.5 is a preferred alternative splicing product of PTK7, with a significantly increased proportion in ACC compared with other cancers. Its expression shows potential prognostic value in terms of PFS and DSS in ACC patients. The methylation status of cg20819617 might play a critical role in modulating PTK7 transcription and ENST00000489707.5 expression.

Pharmacological characterization of the 5-HT1A receptor of Bombyx mori and its role in locomotion.

Serotonin is involved in the regulation of many physiological and behavioral processes in vertebrates and invertebrates. The effects of serotonin are mediated through interactions of several 5-HT receptor types. The expression and pharmacological properties of 5-HT1 have received more attention than other serotonin receptors, but its functions at the individual level are little studied in arthropods. Silkworm, a Lepidoptera model, almost has no reports about serotonin receptors. To analyze the function of Bm5-HT1A receptor in vitro, the ORF of Bm5-HT1A was cloned into the pcDNA3.1 vector and expressed in HEK 293 cells. Serotonin activation of Bm5-HT1A-expressing cells decreased forskolin-stimulated cAMP synthesis and had the most potent effect compared to other biogenic amines. Serotonin reduced cAMP synthesis in a dose-dependent manner, and half-maximal activation (EC50) occurred at a concentration of 1.17 × 10-7 M (117 nM). The pharmacological analysis demonstrated that the rank potency of agonists was pimozide >8-OH-DPAT >5-MeOT ~ αm-5-HT, and antagonists was WAY-100635 > prazosin > SB-269970 > methiothepin at the Bm5-HT1A receptor. Injecting the antagonist of Bm5-HT1A receptor into larvae caused slow or weak motility, and adults lowered courtship vitality or moving speed. Injecting dsRNA of Bm5-HT1A into adults also dropped locomotivity in courtship. These results show that the Bm5-HT1A receptor is related to locomotor activity. This study provides the first information of serotonin receptor on pharmacological in silkworm and on individual functions in arthropods.

PTK7 is a novel oncogenic target for esophageal squamous cell carcinoma.

BACKGROUND: Overexpression of PTK7 has been found in multiple cancers and has been proposed to serve as a prognostic marker for intrahepatic cholangiocarcinoma. Its role in esophageal cancer, however, remains to be clarified. We hypothesize that PTK7 positively regulates tumorigenesis of esophageal cancer. METHODS: We examined PTK7 expression pattern in human esophageal squamous carcinoma by Oncomine expression analysis and by immunohistochemistry (IHC) staining. We knocked down PTK7 in two esophageal squamous cell carcinoma cell lines, TE-5, and TE-9, by siRNA, and evaluated cell proliferation, apoptosis, and migration ofPTK7-defective cells. Expressions of major apoptotic regulators and effectors were also determined by quantitative real-time PCR in PTK7-defective cells. We further overexpressed PTK7 in the cell to evaluate its effects on cell proliferation, apoptosis, and migration. RESULTS: Both Oncomine expression and IHC analyses showed that PTK7 is overexpressed in clinical esophageal squamous cell carcinoma tumors. PTK7 siRNA suppressed cell growth and promoted apoptosis of TE-5 and TE-9. PTK7-defective cells further displayed reduced cellular migration that was concomitant with upregulation of E-cadherin. Conversely, overexpression of PTK7 promotes cell proliferation and invasion, while apoptosis of the PTK7-overexpressing cells is repressed. Notably, major apoptotic regulators, such as p53 and caspases, are significantly upregulated in siPTK7 cells. CONCLUSIONS: PTK7 plays an oncogenic role in tumorigenesis and metastasis of esophageal squamous carcinoma. PTK7 achieves its oncogenic function in esophageal squamous cell carcinoma partially through the negative regulation of apoptosis.

Balance maintenance in high-speed motion of humanoid robot arm-based on the 6D constraints of momentum change rate.

Based on the 6D constraints of momentum change rate (CMCR), this paper puts forward a real-time and full balance maintenance method for the humanoid robot during high-speed movement of its 7-DOF arm. First, the total momentum formula for the robot's two arms is given and the momentum change rate is defined by the time derivative of the total momentum. The author also illustrates the idea of full balance maintenance and analyzes the physical meaning of 6D CMCR and its fundamental relation to full balance maintenance. Moreover, discretization and optimization solution of CMCR has been provided with the motion constraint of the auxiliary arm's joint, and the solving algorithm is optimized. The simulation results have shown the validity and generality of the proposed method on the full balance maintenance in the 6 DOFs of the robot body under 6D CMCR. This method ensures 6D dynamics balance performance and increases abundant ZMP stability margin. The resulting motion of the auxiliary arm has large abundance in joint space, and the angular velocity and the angular acceleration of these joints lie within the predefined limits. The proposed algorithm also has good real-time performance.

Generation and characterization of mAb 61H9 against junctional adhesion molecule-a with potent antitumor activity.

Junctional adhesion molecule-A (JAM-A) is an adhesion molecule that exists on the surface of certain types of cells, including white blood cells, endothelial cells, and dendritic cells. In this study, the cDNA sequences of JAM-A-Fc were chemically synthesized with optimization for mammalian expression. Afterward, we analyzed JAM-A protein expression through transient transfection in HEK293 cell lines. Mice were immunized with JAM-A-Fc protein, and hybridoma was prepared by fusing myeloma cells and mouse spleen cells. Antibodies were purified from the hybridoma supernatant and four monoclonal strains were obtained and numbered 61H9, 70E5, 71A8, and 74H3 via enzyme-linked immunosorbent assay screening. Immunofluorescence staining assay showed 61H9 was the most suitable cell line for mAb production due to its fluorescence signal being the strongest. Flow cytometric analysis proved that 61H9 possessed high affinity. Moreover, antagonism of JAM-A mAb could attenuate the proliferative, migrative, and invasive abilities of ESCC cells and significantly inhibit tumor growth in mice. By examining hematoxylin-eosin staining mice tumor tissues, we found inflammatory cells infiltrated lightly in the anti-JAM-A group. The expression of BCL-2 and IκBα in the anti-JAM-A group were decreased in mice tumor tissues compared to the control group. Ultimately, a method for preparing high-yield JAM-A-Fc protein was created and a high affinity mAb against JAM-A with an antitumor effect was prepared.

Drosophila Larvae-Inspired Soft Crawling Robot with Multimodal Locomotion and Versatile Applications.

Soft crawling robots have been widely studied and applied because of their excellent environmental adaptability and flexible movement. However, most existing soft crawling robots typically exhibit a single-motion mode and lack diverse capabilities. Inspired by Drosophila larvae, this paper proposes a compact soft crawling robot (weight, 13 g; length, 165 mm; diameter, 35 mm) with multimodal locomotion (forward, turning, rolling, and twisting). Each robot module uses 4 sets of high-power-density shape memory alloy actuators, endowing it with 4 degrees of motion freedom. We analyze the mechanical characteristics of the robot modules through experiments and simulation analysis. The plug-and-play modules can be quickly assembled to meet different motion and task requirements. The soft crawling robot can be remotely operated with an external controller, showcasing multimodal motion on various material surfaces. In a narrow maze, the robot demonstrates agile movement and effective maneuvering around obstacles. In addition, leveraging the inherent bistable characteristics of the robot modules, we used the robot modules as anchoring units and installed a microcamera on the robot's head for pipeline detection. The robot completed the inspection in horizontal, vertical, curved, and branched pipelines, adjusted the camera view, and twisted a valve in the pipeline for the first time. Our research highlights the robot's superior locomotion and application capabilities, providing an innovative strategy for the development of lightweight, compact, and multifunctional soft crawling robots.

Unified Cross-Structural Motion Retargeting for Humanoid Characters.

Motion retargeting for animation characters has potential applications in fields such as animation production and virtual reality. However, current methods either assume that the source and target characters have the same skeletal structure, or require designing and training specific model architectures for each structure. In this paper, we aim to address the challenge of motion retargeting across previously unseen skeletal structures with a unified dynamic graph network. The proposed approach utilizes a dynamic graph transformation module to dynamically transfer latent motion features to different structures. We also take into consideration for intricate hand movements and model both torso and hand joints as graphs in a unified manner for whole-body motion retargeting. Our model allows the use of motion data from different structures to train a unified model and learns cross-structural motion retargeting in an unsupervised manner with unpaired data. Experimental results demonstrate the superiority of the proposed method in terms of data efficiency and performance on both seen and unseen structures.

A Review of Intelligent Walking Support Robots: Aiding Sit-to-Stand Transition and Walking.

Nowadays, numerous countries are facing the challenge of aging population. Additionally, the number of people with reduced mobility due to physical illness is increasing. In response to this issue, robots used for walking assistance and sit-to-stand (STS) transition have been introduced in nursing to assist these individuals with walking. Given the shared characteristics of these robots, this paper collectively refers to them as Walking Support Robots (WSR). Additionally, service robots with assisting functions have been included in the scope of this review. WSR are a crucial element of modern nursing assistants and have received significant research attention. Unlike passive walkers that require much user's strength to move, WSR can autonomously perceive the state of the user and environment, and select appropriate control strategies to assist the user in maintaining balance and movement. This paper offers a comprehensive review of recent literature on WSR, encompassing an analysis of structure design, perception methods, control strategies and safety & comfort features. In conclusion, it summarizes the key findings, current challenges and discusses potential future research directions in this field.

AI co-pilot bronchoscope robot.

The unequal distribution of medical resources and scarcity of experienced practitioners confine access to bronchoscopy primarily to well-equipped hospitals in developed regions, contributing to the unavailability of bronchoscopic services in underdeveloped areas. Here, we present an artificial intelligence (AI) co-pilot bronchoscope robot that empowers novice doctors to conduct lung examinations as safely and adeptly as experienced colleagues. The system features a user-friendly, plug-and-play catheter, devised for robot-assisted steering, facilitating access to bronchi beyond the fifth generation in average adult patients. Drawing upon historical bronchoscopic videos and expert imitation, our AI-human shared control algorithm enables novice doctors to achieve safe steering in the lung, mitigating misoperations. Both in vitro and in vivo results underscore that our system equips novice doctors with the skills to perform lung examinations as expertly as seasoned practitioners. This study offers innovative strategies to address the pressing issue of medical resource disparities through AI assistance.

A revision of the genus Trichohoplorana Breuning, 1961 (Arthropoda, Insecta, Coleoptera, Cerambycidae, Lamiinae, Acanthocinini).

A taxonomic revision of the genus Trichohoplorana Breuning, 1961 is presented. A junior synonym of Trichohoplorana, Ipochiromima Sama & Sudre, 2009, syn. nov., is proposed. A junior synonym of T.dureli Breuning, 1961, I.sikkimensis (Breuning, 1982), syn. nov., is proposed. Trichohoplorana is newly recorded from Vietnam. A new species, T.nigeralbasp. nov. is described from Vietnam. Trichohoploranaluteomaculata Gouverneur, 2016 is newly recorded from China and Vietnam. Hind wings and male terminalia of T.luteomaculata are described for the first time. Trichohoplorana is redescribed, and a key to Trichohoplorana species is presented.

Possible Regulation of Larval Juvenile Hormone Titers in Bombyx mori by BmFAMeT6.

Juvenile hormone (JH) plays a vital role in the growth, development, and reproduction of insects and other arthropods. Previous experiments have suggested that BmFAMeT6 could affect the duration of the silk moth's larval stage. In this study, we established the BmFAMeT6 overexpression strain and BmFAMeT6 knockout strain using the GAL4/UAS binary hybrid system and CRISPR/Cas 9 system, respectively, and found that the larval stage of the overexpression strain was shorter, while the knockout strain was longer. Our results exhibited that both the JH titers and BmKr-h1 levels in the larvae of the third instar were reduced significantly by BmFAMeT6 overexpression, but were increased obviously by BmFAMeT6 knockout. In addition, injection of farnesoic acid induced changes in the JH I and JH II levels in the hemolymphs of larvae. This study is the first to directly reveal the role of BmFAMeT6 in the regulation of insect JH titers and the relationship between farnesoic acid and JH (JH I and JH II). This provides a new perspective on regulating the growth and development of insects such as Bombyx mori.

Effect of peach trichome removal on post-harvest brown rot and on the fruit surface microbiome.

Postharvest peaches undergo rapid soft ripening and are susceptible to fungal diseases, which often result in severe losses during storage. The peach epidermis contains trichomes that form a specific structure on the peach surface. However, the relationship between trichomes and postharvest disease and involved mechanisms has not been well studied. In this study, the removal of trichomes reduced the disease incidence of peach brown rot caused by Monilinia fructicola. Cryo-scanning electron microscope observations showed that the fungal hyphae were found attached to the surface of trichomes. The fungal and bacterial communities on the peach surface at 0 d and 6 d were obtained by amplicon sequencing technology. Fungal communities on the peach surface contained a total of 1089 amplicon sequence variants (ASVs), which were demarcated into eight fungal phyla, 25 classes, 66 orders, 137 families, and 228 genera. The bacterial communities contained 10,821 ASVs assigned to 25 phyla, 50 classes, 114 orders, 220 families, and 507 genera. Higher bacterial diversity than fungal diversity was recorded on the peach epidermis. Trichome removal changed the microbial diversity and community on the peach surface. Compared with peach epidermis samples, the peach epidermis excluded trichomes samples contained similar fungal alpha diversity but significantly lower bacterial diversity. Seventeen different fungal genera and twenty-eight different bacterial genera were identified between peach trichome and peach epidermis excluded trichomes samples. The fungal and bacterial diversity on the peach epidermis showed a decreasing trend during storage. Beta diversity analysis revealed that the microbial communities of the peach epidermis and trichomes show different change trends between 0 d and 6 d. Trichome removal decreased relative abundance of Monilinia spp. and increased relative abundance of potential yeast and bacterial biocontrol agents. This study suggested that trichomes might modulate the microbial communities on fruit surfaces, and trichome removal technology after harvest might be developed to control peach postharvest decay.

DPCN++: Differentiable Phase Correlation Network for Versatile Pose Registration.

Pose registration is critical in vision and robotics. This article focuses on the challenging task of initialization-free pose registration up to 7DoF for homogeneous and heterogeneous measurements. While recent learning-based methods show promise using differentiable solvers, they either rely on heuristically defined correspondences or require initialization. Phase correlation seeks solutions in the spectral domain and is correspondence-free and initialization-free. Following this, we propose a differentiable solver and combine it with simple feature extraction networks, namely DPCN++. It can perform registration for homo/hetero inputs and generalizes well on unseen objects. Specifically, the feature extraction networks first learn dense feature grids from a pair of homogeneous/heterogeneous measurements. These feature grids are then transformed into a translation and scale invariant spectrum representation based on Fourier transform and spherical radial aggregation, decoupling translation and scale from rotation. Next, the rotation, scale, and translation are independently and efficiently estimated in the spectrum step-by-step. The entire pipeline is differentiable and trained end-to-end. We evaluate DCPN++ on a wide range of tasks taking different input modalities, including 2D bird's-eye view images, 3D object and scene measurements, and medical images. Experimental results demonstrate that DCPN++ outperforms both classical and learning-based baselines, especially on partially observed and heterogeneous measurements.

Weakly-Interactive-Mixed Learning: Less Labelling Cost for Better Medical Image Segmentation.

Common medical image segmentation tasks require large training datasets with pixel-level annotations which are very expensive and time-consuming to prepare. To overcome such limitation and achieve the desired segmentation accuracy, a novel Weakly-Interactive-Mixed Learning (WIML) framework is proposed by efficiently using weak labels. On one hand, utilize weak labels to reduce annotation time for high-quality strong labels by designing a Weakly-Interactive Annotation (WIA) part of the WIML which prudently introduces interactive learning into the weakly-supervised segmentation strategy. On the other hand, utilize weak labels and very few strong labels to achieve desired segmentation accuracy by designing a Mixed-Supervised Learning (MSL) part of the WIML which can boost the segmentation accuracy by providing strong prior knowledge during training. Besides, a multi-task Full-Parameter-Sharing Network (FPSNet) is proposed to better implement this framework. Specifically, to further reduce annotation time, attention modules (scSE) are integrated into FPSNet to improve the class activation map (CAM) performance for the first time. To further improve segmentation accuracy, a Full-Parameter-Sharing (FPS) strategy is designed in FPSNet to alleviate the overfitting of the segmentation task supervised by very few strong labels. The proposed method is validated on the BraTS 2019 and LiTS 2017 datasets, and experiments demonstrate that the proposed method WIML-FPSNet outperforms several state-of-the-art segmentation methods with minimal annotation efforts.

Learning a simulation-based visual policy for real-world peg in unseen holes.

This paper proposes a learning-based visual peg-in-hole that enables training with several shapes in simulation and adapting to arbitrary unseen shapes in the real world with minimal sim-to-real cost. The core idea is to decouple the generalization of the sensory-motor policy from the design of a fast-adaptable perception module and a simulated generic policy module. The framework consists of a segmentation network (SN), a virtual sensor network (VSN), and a controller network (CN). Concretely, the VSN is trained to measure the pose of the unseen shape from a segmented image. After that, given the shape-agnostic pose measurement, the CN is trained to achieve a generic peg-in-hole. Finally, when applying to real unseen holes, we only have to fine-tune the SN required by the simulated VSN + CN. To further minimize the transfer cost, we propose to automatically collect and annotate the data for the SN after one-minute human teaching. Simulated and real-world results are presented under the configuration of eye-to/in-hand. An electric vehicle charging system with the proposed policy inside achieves a 10/10 success rate in 2-3 s, using only hundreds of auto-labeled samples for the SN transfer.