Heterogeneous Structure Omnidirectional Strain Sensor Arrays With Cognitively Learned Neural Networks.

Mechanically stretchable strain sensors gain tremendous attention for bioinspired skin sensation systems and artificially intelligent tactile sensors. However, high-accuracy detection of both strain intensity and direction with simple device/array structures is still insufficient. To overcome this limitation, an omnidirectional strain perception platform utilizing a stretchable strain sensor array with triangular-sensor-assembly (three sensors tilted by 45°) coupled with machine learning (ML) -based neural network classification algorithm, is proposed. The strain sensor, which is constructed with strain-insensitive electrode regions and strain-sensitive channel region, can minimize the undesirable electrical intrusion from the electrodes by strain, leading to a heterogeneous surface structure for more reliable strain sensing characteristics. The strain sensor exhibits decent sensitivity with gauge factor (GF) of ≈8, a moderate sensing range (≈0-35%), and relatively good reliability (3000 stretching cycles). More importantly, by employing a multiclass-multioutput behavior-learned cognition algorithm, the stretchable sensor array with triangular-sensor-assembly exhibits highly accurate recognition of both direction and intensity of an arbitrary strain by interpretating the correlated signals from the three-unit sensors. The omnidirectional strain perception platform with its neural network algorithm exhibits overall strain intensity and direction accuracy around 98% ± 2% over a strain range of ≈0-30% in various surface stimuli environments.

SSD-EMB: An Improved SSD Using Enhanced Feature Map Block for Object Detection.

The development of deep learning has achieved great success in object detection, but small object detection is still a difficult and challenging task in computer vision. To address the problem, we propose an improved single-shot multibox detector (SSD) using enhanced feature map blocks (SSD-EMB). The enhanced feature map block (EMB) consists of attention stream and feature map concatenation stream. The attention stream allows the proposed model to focus on the object regions rather than background owing to channel averaging and the effectiveness of the normalization. The feature map concatenation stream provides additional semantic information to the model without degrading the detection speed. By combining the output of these two streams, the enhanced feature map, which improves the detection of a small object, is generated. Experimental results show that the proposed model has high accuracy in small object detection. The proposed model not only achieves good detection accuracy, but also has a good detection speed. The SSD-EMB achieved a mean average precision (mAP) of 80.4% on the PASCAL VOC 2007 dataset at 30 frames per second on an RTX 2080Ti graphics processing unit, an mAP of 79.9% on the VOC 2012 dataset, and an mAP of 26.6% on the MS COCO dataset.

Suppression of ARID1A associated with decreased CD8 T cells improves cell survival of ovarian clear cell carcinoma.

OBJECTIVE: AT-rich interactive domain 1A (ARID1A) plays an important role as a tumor suppressor gene in ovarian clear cell carcinoma (OCCC), but the clinical application of ARID1A remains unclear. The aim of this study was to analyze clinicopathological parameters, molecular interactions and immune-infiltration in patients with low ARID1A expression and to provide candidate target drugs. METHODS: We investigated the clinicopathologic parameters, specific gene sets/genes, and immunological relevance according to ARID1A expression in 998 OCCC patients from 12 eligible studies (using meta-analyses); 30 OCCC patients from the Hanyang University Guri Hospital (HYGH) cohort; and 52 OCCC patients from gene set enrichment (GSE) 65986 (25 patients), 63885 (9 patients), and 54809 (6 patients and 12 healthy people) of the Gene Expression Omnibus (GEO). We analyzed network-based pathways based on gene set enrichment analysis (GSEA) and performed in vitro drug screening. RESULTS: Low ARID1A expression was associated with poor survival in OCCC from the meta-analysis, HYGH cohort and GEO data. In GSEA, low ARID1A expression was related to the tumor invasion process as well as a low immune-infiltration. In silico cytometry showed that CD8 T cells were decreased with low ARID1A expression. In pathway analysis, ARID1A was associated with angiogenic endothelial cell signaling. In vitro drug screening revealed that cabozantinib and bicalutamide effectively inhibited specific hub genes, such as vascular endothelial growth factor-A and androgen receptor, in OCCC cells with low ARID1A expression. CONCLUSIONS: Therapeutic strategies making use of low ARID1A could contribute to better clinical management/research for patients with OCCC.

SSD-TSEFFM: New SSD Using Trident Feature and Squeeze and Extraction Feature Fusion.

The single shot multi-box detector (SSD) exhibits low accuracy in small-object detection; this is because it does not consider the scale contextual information between its layers, and the shallow layers lack adequate semantic information. To improve the accuracy of the original SSD, this paper proposes a new single shot multi-box detector using trident feature and squeeze and extraction feature fusion (SSD-TSEFFM); this detector employs the trident network and the squeeze and excitation feature fusion module. Furthermore, a trident feature module (TFM) is developed, inspired by the trident network, to consider the scale contextual information. The use of this module makes the proposed model robust to scale changes owing to the application of dilated convolution. Further, the squeeze and excitation block feature fusion module (SEFFM) is used to provide more semantic information to the model. The SSD-TSEFFM is compared with the faster regions with convolution neural network features (RCNN) (2015), SSD (2016), and DF-SSD (2020) on the PASCAL VOC 2007 and 2012 datasets. The experimental results demonstrate the high accuracy of the proposed model in small-object detection, in addition to a good overall accuracy. The SSD-TSEFFM achieved 80.4% mAP and 80.2% mAP on the 2007 and 2012 datasets, respectively. This indicates an average improvement of approximately 2% over other models.

A Behavior-Learned Cross-Reactive Sensor Matrix for Intelligent Skin Perception.

Mimicking human skin sensation such as spontaneous multimodal perception and identification/discrimination of intermixed stimuli is severely hindered by the difficulty of efficient integration of complex cutaneous receptor-emulating circuitry and the lack of an appropriate protocol to discern the intermixed signals. Here, a highly stretchable cross-reactive sensor matrix is demonstrated, which can detect, classify, and discriminate various intermixed tactile and thermal stimuli using a machine-learning approach. Particularly, the multimodal perception ability is achieved by utilizing a learning algorithm based on the bag-of-words (BoW) model, where, by learning and recognizing the stimulus-dependent 2D output image patterns, the discrimination of each stimulus in various multimodal stimuli environments is possible. In addition, the single sensor device integrated in the cross-reactive sensor matrix exhibits multimodal detection of strain, flexion, pressure, and temperature. It is hoped that his proof-of-concept device with machine-learning-based approach will provide a versatile route to simplify the electronic skin systems with reduced architecture complexity and adaptability to various environments beyond the limitation of conventional "lock and key" approaches.

Dilated Skip Convolution for Facial Landmark Detection.

Facial landmark detection has gained enormous interest for face-related applications due to its success in facial analysis tasks such as facial recognition, cartoon generation, face tracking and facial expression analysis. Many studies have been proposed and implemented to deal with the challenging problems of localizing facial landmarks from given images, including large appearance variations and partial occlusion. Studies have differed in the way they use the facial appearances and shape information of input images. In our work, we consider facial information within both global and local contexts. We aim to obtain local pixel-level accuracy for local-context information in the first stage and integrate this with knowledge of spatial relationships between each key point in a whole image for global-context information in the second stage. Thus, the pipeline of our architecture consists of two main components: (1) a deep network for local-context subnet that generates detection heatmaps via fully convolutional DenseNets with additional kernel convolution filters and (2) a dilated skip convolution subnet-a combination of dilated convolutions and skip-connections networks-that are in charge of robustly refining the local appearance heatmaps. Through this proposed architecture, we demonstrate that our approach achieves state-of-the-art performance on challenging datasets-including LFPW, HELEN, 300W and AFLW2000-3D-by leveraging fully convolutional DenseNets, skip-connections and dilated convolution architecture without further post-processing.

Energy-Efficient Cluster-Head Selection for Wireless Sensor Networks Using Sampling-Based Spider Monkey Optimization.

Extending the lifetime and stability of wireless sensor networks (WSNs) through efficient energy consumption remains challenging. Though clustering has improved energy efficiency through cluster-head selection, its application is still complicated. In existing cluster-head selection methods, the locations where cluster-heads are desirable are first searched. Next, the nodes closest to these locations are selected as the cluster-heads. This location-based approach causes problems such as increased computation, poor selection accuracy, and the selection of duplicate nodes. To solve these problems, we propose the sampling-based spider monkey optimization (SMO) method. If the sampling population consists of nodes to select cluster-heads, the cluster-heads are selected among the nodes. Thus, the problems caused by different locations of nodes and cluster-heads are resolved. Consequently, we improve lifetime and stability of WSNs through sampling-based spider monkey optimization and energy-efficient cluster head selection (SSMOECHS). This study describes how the sampling method is used in basic SMO and how to select cluster-heads using sampling-based SMO. The experimental results are compared to similar protocols, namely low-energy adaptive clustering hierarchy centralized (LEACH-C), particle swarm optimization clustering protocol (PSO-C), and SMO based threshold-sensitive energy-efficient delay-aware routing protocol (SMOTECP), and the results are shown in both homogeneous and heterogeneous setups. In these setups, SSMOECHS improves network lifetime and stability periods by averages of 13.4%, 7.1%, 34.6%, and 1.8%, respectively.

Frequency-Stable Ionic-Type Hybrid Gate Dielectrics for High Mobility Solution-Processed Metal-Oxide Thin-Film Transistors.

In this paper, we demonstrate high mobility solution-processed metal-oxide thin-film transistors (TFTs) by using a high-frequency-stable ionic-type hybrid gate dielectric (HGD). The HGD gate dielectric, a blend of sol-gel aluminum oxide (AlOx) and poly(4-vinylphenol) (PVP), exhibited high dielectric constant (ε~8.15) and high-frequency-stable characteristics (1 MHz). Using the ionic-type HGD as a gate dielectric layer, an minimal electron-double-layer (EDL) can be formed at the gate dielectric/InOx interface, enhancing the field-effect mobility of the TFTs. Particularly, using the ionic-type HGD gate dielectrics annealed at 350 °C, InOx TFTs having an average field-effect mobility of 16.1 cm²/Vs were achieved (maximum mobility of 24 cm²/Vs). Furthermore, the ionic-type HGD gate dielectrics can be processed at a low temperature of 150 °C, which may enable their applications in low-thermal-budget plastic and elastomeric substrates. In addition, we systematically studied the operational stability of the InOx TFTs using the HGD gate dielectric, and it was observed that the HGD gate dielectric effectively suppressed the negative threshold voltage shift during the negative-illumination-bias stress possibly owing to the recombination of hole carriers injected in the gate dielectric with the negatively charged ionic species in the HGD gate dielectric.

Comparative proteomic analysis of malformed umbilical cords from somatic cell nuclear transfer-derived piglets: implications for early postnatal death.

BACKGROUND: Somatic cell nuclear transfer (scNT)-derived piglets have high rates of mortality, including stillbirth and postnatal death. Here, we examined severe malformed umbilical cords (MUC), as well as other organs, from nine scNT-derived term piglets. RESULTS: Microscopic analysis revealed complete occlusive thrombi and the absence of columnar epithelial layers in MUC (scNT-MUC) derived from scNT piglets. scNT-MUC had significantly lower expression levels of platelet endothelial cell adhesion molecule-1 (PECAM-1) and angiogenesis-related genes than umbilical cords of normal scNT piglets (scNT-N) that survived into adulthood. Endothelial cells derived from scNT-MUC migrated and formed tubules more slowly than endothelial cells from control umbilical cords or scNT-N. Proteomic analysis of scNT-MUC revealed significant down-regulation of proteins involved in the prevention of oxidative stress and the regulation of glycolysis and cell motility, while molecules involved in apoptosis were significantly up-regulated. Histomorphometric analysis revealed severe calcification in the kidneys and placenta, peliosis in the liver sinusoidal space, abnormal stromal cell proliferation in the lungs, and tubular degeneration in the kidneys in scNT piglets with MUC. Increased levels of apoptosis were also detected in organs derived from all scNT piglets with MUC. CONCLUSION: These results suggest that MUC contribute to fetal malformations, preterm birth and low birth weight due to underlying molecular defects that result in hypoplastic umbilical arteries and/or placental insufficiency. The results of the current study demonstrate the effects of MUC on fetal growth and organ development in scNT-derived pigs, and provide important insight into the molecular mechanisms underlying angiogenesis during umbilical cord development.