Recognition Step Research Articles

Down-regulation of LINC-ROR, HOXA-AS2 and MEG3 in gastric cancer

Long non-coding RNAs (lncRNAs) have been identified as modulators of gastric carcinogenesis. Evaluation of expression amounts of these transcripts is a primary but essential step for recognition of the role of lncRNAs in the carcinogenesis. Therefore, we compared expressions of LINC-ROR, HOXA-AS2, MEG3 and HOTTIP lncRNAs in gastric cancer samples and nearby non-cancerous samples. Expression levels of LINC-ROR, HOXA-AS2 and MEG3 lncRNAs have been lower in gastric cancer samples compared with nearby non-cancerous samples (Expression ratios = 0.26, 0.37 and 0.36; P values = 0.021, 0.015 and 0.032, respectively). However, expression levels of HOTTIP were not significantly different between gastric cancer tissues and nearby tissues (P value = 0.43). HOTTIP expression was associated with tumor size (P value = 0.04). In addition, MEG3 expression was associated with site of primary tumor (P = 0.0003). Expressions of LINC-ROR and HOXA-AS2 were not associated with any clinical or pathological parameter. ROC curve analysis revealed that HOXA-AS2 and LINC-ROR could significantly differentiate between gastric cancer samples and nearby non-cancerous tissues (AUC values = 0.68 and 0.64; P values = 0.01 and 0.04, respectively). Taken together, the current investigation provides clues for contribution of LINC-ROR, HOXA-AS2 and MEG3 lncRNAs in gastric carcinogenesis and warrants further mechanistical assays.

DACTYLOLOGY RECOGNITION SYSTEM

Dactylology is the technique of communicating with signs made with fingers, especially in the manual alphabets used by the deaf.Sign Language is one of the oldest and most natural form of language for interaction between normal and deaf & mute people, but since most people do not know sign language and interpreters are very difficult to find, we have come up with a real time method using neural networks for fingerspelling based American sign Language. Sign language is a visual language and consists of 3 major components: Fingerspelling, Word level vocabulary and Facial expressions. In our project we basically focus on producing a model which can recognise Fingerspelling based hand gestures in order to form a complete word by combining each gesture. We follow 4 steps in the hand gesture recognition: i) Data Acquisition, ii) Data preprocessing, iii) Feature extraction iv) Gesture classification. We aim to collect 600 RGB images each for 26 alphabets and apply a filter called Gaussian blur to each of the images. This filter gives a black& white outline of the hand gesture which makes it easier for computation. The images are then passed through convolution neural network classifier which predicts the class of the hand gestures. Our method provides 95% accuracy for the 26 letters of the alphabet

Use Brain-Like Audio Features to Improve Speech Recognition Performance

Speech recognition plays an important role in the field of human-computer interaction through the use of acoustic sensors, but speech recognition is technically difficult, has complex overall logic, relies heavily on neural network algorithms, and has extremely high technical requirements. In speech recognition, feature extraction is the first step in speech recognition for recovering and extracting speech features. Existing methods, such as Meier spectral coefficients (MFCC) and spectrograms, lose a large amount of acoustic information and lack biological interpretability. Then, for example, existing speech self-supervised representation learning methods based on contrast prediction need to construct a large number of negative samples during training, and their learning effects depend on large batches of training, which requires a large amount of computational resources for the problem. Therefore, in this paper, we propose a new feature extraction method, called SHH (spike-H), that resembles the human brain and achieves higher speech recognition rates than previous methods. The features extracted using the proposed model are subsequently fed into the classification model. We propose a novel parallel CRNN model with an attention mechanism that considers both temporal and spatial features. Experimental results show that the proposed CRNN achieves an accuracy of 94.8% on the Aurora dataset. In addition, audio similarity experiments show that SHH can better distinguish audio features. In addition, the ablation experiments show that SHH is applicable to digital speech recognition.

Multi-modal cyber security based object detection by classification using deep learning and background suppression techniques

Foreground object recognition in video is critical in many computer vision applications and automated video surveillance systems. Object detection and tracking are critical steps in navigation, object recognition and surveillance schemes. Object detection is process of separating foreground and background items in photographs. In this paper, we proposes a framework for achieving these tasks with the enhanced cyber security control facility. This proposed algorithmperformed in 2 stages: multi-object detection utilizing the Cyber secure Probabilistic Gaussian Mixture Model (Cy_SPGMM) and background suppression and another stage is multiple moving objects tracking utilizing Kernel convoluted moving window with Kalman filter (KCMW_KF). It can, however, deal with a variety of video sequences in the MOT 20 dataset. The experimental findings reveal that the proposed algorithm detects and tracks foreground objects in complex and dynamic scenarios with high accuracy, robustness, and efficiency. This method also produces smoothened images without noise.

Promoting Tumor Accumulation of Anticancer Drugs by Hierarchical Carrying of Exogenous and Endogenous Vehicles

Promoting tumor accumulation of active pharmaceutical ingredients is crucial for targeted anticancer therapy. However, this issue has not yet been addressed because of the fast clearance or premature degradation, slow drug release kinetics, and nonspecific biodistribution of the reported formulations. Herein, a new drug delivery paradigm is proposed based on supramolecular hierarchical recognition to achieve high tumor accumulation of anticancer drugs by overcoming these three challenges. The prepared supramolecular ternary formulation of paclitaxel (PTX) contains two steps of molecular recognition: exogenous recognition of PTX by an artificial macrocyclic carrier, sulfonated azocalix[5]arene (SAC5A) in vitro, and endogenous recognition of PTX@SAC5A by intrinsic serum albumin in vivo. The ternary PTX@SAC5A⊂albumin concurrently accomplishes prolonged blood circulation, rapid drug release, and dual passive and hypoxia‐responsive targeting. As a result, the PTX@SAC5A⊂albumin formulation exhibits more efficient tumor accumulation than free PTX and albumin‐based, solvent‐based, liposome‐based, and sulfobutyl ether‐β‐cyclodextrin‐based PTX formulations, thereby contributing to better therapeutic efficacy. The current strategy paves the way for promoting the tumor accumulation of drugs, showing the advantages of simplicity, universality, and reproducibility.

Therapeutic Strategies to Enhance Tumor Antigenicity: Making the Tumor Detectable by the Immune System.

Cancer immunotherapy has revolutionized the oncology field, but many patients still do not respond to current immunotherapy approaches. One of the main challenges in broadening the range of responses to this type of treatment is the limited source of tumor neoantigens. T cells constitute a main line of defense against cancer, and the decisive step to trigger their activation is mediated by antigen recognition. Antigens allow the immune system to differentiate between self and foreign, which constitutes a critical step in recognition of cancer cells and the consequent development or control of the malignancy. One of the keystones to achieving a successful antitumor response is the presence of potent tumor antigens, known as neoantigens. However, tumors develop strategies to evade the immune system and resist current immunotherapies, and many tumors present a low tumor mutation burden limiting the presence of tumor antigenicity. Therefore, new approaches must be taken into consideration to overcome these shortcomings. The possibility of making tumors more antigenic represents a promising front to further improve the success of immunotherapy in cancer. Throughout this review, we explored different state-of-the-art tools to induce the presentation of new tumor antigens by intervening at protein, mRNA or genomic levels in malignant cells.

XAB2 dynamics during DNA damage-dependent transcription inhibition.

Xeroderma Pigmentosum group A-binding protein 2 (XAB2) is a multifunctional protein playing a critical role in distinct cellular processes including transcription, splicing, DNA repair, and messenger RNA export. In this study, we demonstrate that XAB2 is involved specifically and exclusively in Transcription-Coupled Nucleotide Excision Repair (TC-NER) reactions and solely for RNA polymerase 2 (RNAP2)-transcribed genes. Surprisingly, contrary to all the other NER proteins studied so far, XAB2 does not accumulate on the local UV-C damage; on the contrary, it becomes more mobile after damage induction. XAB2 mobility is restored when DNA repair reactions are completed. By scrutinizing from which cellular complex/partner/structure XAB2 is released, we have identified that XAB2 is detached after DNA damage induction from DNA:RNA hybrids, commonly known as R-loops, and from the CSA and XPG proteins. This release contributes to the DNA damage recognition step during TC-NER, as in the absence of XAB2, RNAP2 is blocked longer on UV lesions. Moreover, we also demonstrate that XAB2 has a role in retaining RNAP2 on its substrate without any DNA damage.

Restricted-Area Adversarial Example Attack for Image Captioning Model

Deep neural networks provide good performance in the fields of image recognition, speech recognition, and text recognition. For example, recurrent neural networks are used by image captioning models to generate text after an image recognition step, thereby providing captions for the images. The image captioning model first extracts features from the image and generates a representation vector; it then generates the text for the image captions by using the recursive neural network. This model has a weakness, however: it is vulnerable to adversarial examples. In this paper, we propose a method for generating restricted adversarial examples that target image captioning models. By adding a minimal amount of noise just to a specific area of an original sample image, the proposed method creates an adversarial example that remains correctly recognizable to humans yet is misinterpreted by the target model. We evaluated the method’s performance through experiments with the MS COCO dataset and using TensorFlow as the machine learning library. The results show that the proposed method generates a restricted adversarial example that is misinterpreted by the target model while minimizing its distortion from the original sample.

Centrally controlled on-chip criminal face recognition embedded in traffic cameras

This paper evaluates the ability of convolutional networks to solve the problems arising with face classification in a constrained environment. It has the design and implementation of Siamese architecture used for face verification using a single set of the photograph. Because of the intrinsic nature of the problem, computer vision is not only a computer science area of research, but also the object of neuroscientific and psychological studies, mainly because of the general opinion that advances in computer image processing and understanding research will provide insights into how our brain works and vice versa. In the scope of the paper, the training process is closely monitored and we evaluate several practices and parameters as well as their impact on network learning. This paper introduces some novel models for all steps of a face recognition system using embedded computers. In the step of single-shot face recognition, we propose a hybrid model combining Openface artificial neural network combined with siamese network architecture to solve the process efficiently. In the first step, faces detected by Google's mediapipe library will be aligned by Geometric corrections. In this alignment step, we propose a new 2D local texture model based on relative eyes inclination.

Automated Data Digitization System for Vehicle Registration Certificates Using Google Cloud Vision API

This study aims to develop an automated data digitization system for the Thai vehicle registration certificate. It is the first system developed as a web service Application Programming Interface (API), which is essential for any enterprise to increase its business value. Currently, this system is available on “www.carjaidee.com”. The system involves four steps: 1) an embedded frame aligns a document to be correctly recognised in the image acquisition step; 2) sharpening and brightness filtering techniques to enhance image quality are applied in the pre-processing step; 3) the Google Cloud Vision API receives a prompt to proceed in the recognition step; 4) a specific domain dictionary to improve accuracy rate is developed for the post-processing step. This study defines 92 images for the experiment by counting the correct words and terms from the output. The findings suggest that the proposed method, which had an average accuracy of 93.28%, was significantly more accurate than the original method using only the Google Cloud Vision API. However, the system is limited because the dictionaries cannot automatically recognise a new word. In the future, we will explore solutions to this problem using natural language processing techniques. Doi: 10.28991/CEJ-2022-08-07-09 Full Text: PDF

PENYANDANG DISABILITAS BERDAYA MELALUI STRATEGI PEMBERDAYAAN PRECIOUS ONE DI MERUYA UTARA, KECAMATAN KEMBANGAN, JAKARTA BARAT

Persons with disabilities are vulnerable to discrimination, limited accessibility, and underestimated in Indonesia. People with disabilities have a lack of physical or mental condition, which is related to body structure, so that they have difficulty doing a job or acting. People with disabilities often do not get their rights, viewed with a negative stigma, which means that the environment of people with disabilities is still having a lot of problems in providing opportunities, participation or recognition for them. So it is important to provide opportunity, equal access and recognition for them. Therefore, Precious One is here to accommodate people with disabilities to be empowered through the manufacture of handicraft products, as one of the strategic steps in bathing and building recognition for them. This paper is to see the empowerment strategies carried out by Precious One for people with disabilities using qualitative research methods that produce descriptive data with data collection techniques in the form of observation, interviews, documents and reason, which use primary data and secondary data. Based on this paper, a Precious One's empowerment strategy is in the form of placing the right targets, providing education and awareness, building recognition, self-confidence and independence of disabilities, building collaborative networks, building good communication, and optimizing marketing through social media and marketplaces.

Intelligent cataract surgery supervision and evaluation via deep learning

PurposeTo assess the performance of a deep learning (DL) algorithm for evaluating and supervising cataract extraction using phacoemulsification with intraocular lens (IOL) implantation based on cataract surgery (CS) videos. Materials and methodsDeepSurgery was trained using 186 standard CS videos to recognize 12 CS steps and was validated in two datasets that contained 50 and 21 CS videos, respectively. A supervision test including 50 CS videos was used to assess the DeepSurgery guidance and alert function. In addition, a real-time test containing 54 CSs was used to compare the DeepSurgery grading performance to an expert panel and residents. ResultsDeepSurgery achieved stable performance for all 12 recognition steps, including the duration between two pairs of adjacent steps in internal validation with an ACC of 95.06% and external validations with ACCs of 88.77% and 88.34%. DeepSurgery also recognized the chronology of surgical steps and alerted surgeons to order of incorrect steps. Six main steps are automatically and simultaneously quantified during the evaluation process (centesimal system). In a real-time comparative test, the DeepSurgery step recognition performance was robust (ACC of 90.30%). In addition, DeepSurgery and an expert panel achieved comparable performance when assessing the surgical steps (kappa ranged from 0.58 to 0.77). ConclusionsDeepSurgery represents a potential approach to provide a real-time supervision and an objective surgical evaluation system for routine CS and to improve surgical outcomes.

Defects Recognition Algorithm Development from Visual UAV Inspections.

Aircraft maintenance plays a key role in the safety of air transport. One of its most significant procedures is the visual inspection of the aircraft skin for defects. This is mainly carried out manually and involves a high skilled human walking around the aircraft. It is very time consuming, costly, stressful and the outcome heavily depends on the skills of the inspector. In this paper, we propose a two-step process for automating the defect recognition and classification from visual images. The visual inspection can be carried out with the use of an unmanned aerial vehicle (UAV) carrying an image sensor to fully automate the procedure and eliminate any human error. With our proposed method in the first step, we perform the crucial part of recognizing the defect. If a defect is found, the image is fed to an ensemble of classifiers for identifying the type. The classifiers are a combination of different pretrained convolution neural network (CNN) models, which we retrained to fit our problem. For achieving our goal, we created our own dataset with defect images captured from aircrafts during inspection in TUI’s maintenance hangar. The images were preprocessed and used to train different pretrained CNNs with the use of transfer learning. We performed an initial training of 40 different CNN architectures to choose the ones that best fitted our dataset. Then, we chose the best four for fine tuning and further testing. For the first step of defect recognition, the DenseNet201 CNN architecture performed better, with an overall accuracy of 81.82%. For the second step for the defect classification, an ensemble of different CNN models was used. The results show that even with a very small dataset, we can reach an accuracy of around 82% in the defect recognition and even 100% for the classification of the categories of missing or damaged exterior paint and primer and dents.

Density-based Clustering for 3D Stacked Pipe Object Recognition using Directly-given Point Cloud Data on Convolutional Neural Network

One of the most commonly faced tasks in industrial robots is bin picking. Much work has been done in this related topic is about grasping and picking an object from the piled bin but ignoring the recognition step in their pipeline. In this paper, a recognition pipeline for industrial bin picking is proposed. Begin with obtaining point cloud data from different manner of stacking objects there are well separated, well piled, and arbitrary piled. Then followed by segmentation using Density-based Spatial Clustering Application with Noise (DBSCAN) to obtain individual object data. The systems then use Convolutional Neural Network (CNN) that consume raw point cloud data. Performance of the segmentation reaches an impressive result in separating objects and network is evaluated under the varying style of stacking objects and give the result with average Accuracy, Recall, Precision, and F1-Score on 98.72%, 95.45%, 99.39%, and 97.33% respectively. Then the obtained model can be used for multiple objects recognition in one scene.

Functional HLA-C expressing trophoblast spheroids as a model to study placental–maternal immune interactions during human implantation

In healthy couples over half of the conceptions result in failed pregnancy and around 30% of them occur during implantation defining it as a rate-limiting step for the success of native and in vitro fertilization. The understanding of the factors regulating each step of implantation and immune recognition is critical for the pregnancy outcome. Creation of 3D-cell culture models, such as spheroids and organoids, is in the focus of placental tissue engineering in attempt to resemble the in vivo complexity of the maternal-fetal interface and to overcome the need of laboratory animals and human embryos. We constructed stable, reliable, and reproducible trophoblast Sw71 spheroids which are functional independently of the serum level in the culture media. These models resemble the hatched human blastocyst in size, shape and function and are useful for in vitro studies of the in vivo concealed human implantation. Since Sw71 spheroids produce HLA-C, the only classical MHC molecule indispensable for establishment of the immune tolerance and proper human implantation, they are applicable for the evaluation not only of implantation itself but also of maternal-trophoblasts immune interactions. In addition, Sw71-blastocyst-like spheroids are manipulable in low-volume platform, easy to monitor and analyze automatically under treatment with favorable/detrimental factors.

A binary dandelion algorithm using seeding and chaos population strategies for feature selection

Feature selection (FS) is an important pre-processing step in data mining and pattern recognition. It can effectively compress the dimensionality of the feature space to reduce computation time and improve classification performance. The meta-heuristic algorithm-based feature selection method by finding the optimal set of features in the solution space has been widely used. However, this method is prone to trap into local optimality in a sufficiently large solution space. In this paper, we first propose a binary dandelion algorithm (BDA) to improve classification accuracy. In addition, to improve the performance of the algorithm, a binary dandelion algorithm using an improved seeding strategy and chaotic populations (SBDA) is proposed in this paper. Firstly, the strategy of optimizing the seeding radius by using the vibrational function and the historical optimal population increases the complexity of the search process and improves the search performance of the algorithm in the solution space. Secondly, when generating seeds, chaotic populations are generated using chaotic operators, which improves the ability of the algorithm to jump out of the local optimum and improves the stability of the algorithm. In this paper, 15 well-established datasets collected from the UCI machine learning database were adopted to compare four variants of BDA using only chaotic population improvement and in the next experiments, both mechanisms are verified to be effective in improving the performance of the algorithm. In addition, this paper compares the proposed BDA algorithm and SBDA algorithm with eight other classical algorithms. The experimental results show that SBDA can obtain fewer features with higher classification accuracy in most cases.

NeDSeM: Neutrosophy Domain-Based Segmentation Method for Malignant Melanoma Images.

Skin lesion segmentation is the first and indispensable step of malignant melanoma recognition and diagnosis. At present, most of the existing skin lesions segmentation techniques often used traditional methods like optimum thresholding, etc., and deep learning methods like U-net, etc. However, the edges of skin lesions in malignant melanoma images are gradually changed in color, and this change is nonlinear. The existing methods can not effectively distinguish banded edges between lesion areas and healthy skin areas well. Aiming at the uncertainty and fuzziness of banded edges, the neutrosophic set theory is used in this paper which is better than fuzzy theory to deal with banded edge segmentation. Therefore, we proposed a neutrosophy domain-based segmentation method that contains six steps. Firstly, an image is converted into three channels and the pixel matrix of each channel is obtained. Secondly, the pixel matrixes are converted into Neutrosophic Set domain by using the neutrosophic set conversion method to express the uncertainty and fuzziness of banded edges of malignant melanoma images. Thirdly, a new Neutrosophic Entropy model is proposed to combine the three memberships according to some rules by using the transformations in the neutrosophic space to comprehensively express three memberships and highlight the banded edges of the images. Fourthly, the feature augment method is established by the difference of three components. Fifthly, the dilation is used on the neutrosophic entropy matrixes to fill in the noise region. Finally, the image that is represented by transformed matrix is segmented by the Hierarchical Gaussian Mixture Model clustering method to obtain the banded edge of the image. Qualitative and quantitative experiments are performed on malignant melanoma image dataset to evaluate the performance of the NeDSeM method. Compared with some state-of-the-art methods, our method has achieved good results in terms of performance and accuracy.

Structural basis for RNA surveillance by the human nuclear exosome targeting (NEXT) complex

RNA quality control relies on co-factors and adaptors to identify and prepare substrates for degradation by ribonucleases such as the 3' to 5' ribonucleolytic RNA exosome. Here, we determined cryogenic electron microscopy structures of human nuclear exosome targeting (NEXT) complexes bound to RNA that reveal mechanistic insights to substrate recognition and early steps that precede RNA handover to the exosome. The structures illuminate ZCCHC8 as a scaffold, mediating homodimerization while embracing the MTR4 helicase and flexibly anchoring RBM7 to the helicase core. All three subunits collaborate to bind the RNA, with RBM7 and ZCCHC8 surveying sequences upstream of the 3' end to facilitate RNA capture by MTR4. ZCCHC8 obscuresMTR4 surfaces important for RNA binding and extrusion as well as MPP6-dependent recruitment and docking onto the RNA exosome core, interactions that contribute to RNA surveillance by coordinating RNA capture, translocation, and extrusion from the helicase to the exosome for decay.

An Arabidopsis Retention and Splicing complex regulates root and embryo development through pre-mRNA splicing.

Pre-mRNA splicing is an important step in the posttranscriptional processing of transcripts and a key regulator of development. The heterotrimeric retention and splicing (RES) complex plays vital roles in the growth and development of yeast, zebrafish, and humans by mediating pre-mRNA splicing of multiple genes. However, whether the RES complex is conserved in plants and what specific functions it has remain unknown. In this study, we identified Arabidopsis (Arabidopsis thaliana) BUD13 (AtBUD13), GROWTH, DEVELOPMENT AND SPLICING 1 (GDS1), and DAWDLE (DDL) as the counterparts of the yeast RES complex subunits Bud site selection protein 13 (Bud13), U2 snRNP component Snu17 (Snu17), and Pre-mRNA leakage protein 1, respectively. Moreover, we showed that RES is an ancient complex evolutionarily conserved in eukaryotes. GDS1 directly interacts with both AtBUD13 and DDL in nuclear speckles. The BUD13 domain of AtBUD13 and the RNA recognition motif domain of GDS1 are necessary and sufficient for AtBUD13-GDS1 interaction. Mutants of AtBUD13, GDS1, and DDL failed to properly splice multiple genes involved in cell proliferation and showed defects in early embryogenesis and root development. In addition, we found that GDS1 and DDL interact, respectively, with the U2 small nuclear ribonucleoproteins auxiliary factor AtU2AF65B and the NineTeen Complex-related splicing factor SKIP, which are essential for early steps of spliceosome assembly and recognition of splice sites. Altogether, our work reveals that the Arabidopsis RES complex is important for root and early embryo development by modulating pre-mRNA splicing.

Learning-free, divide and conquer text-line extraction algorithm for printed Arabic text with diacritics

The extraction of text lines from document images is a critical step in optical character recognition. It is still considered an open document analysis problem. The presence of numerous font variations, diacritics, overlapping, and touching text-lines presents a challenge to algorithms designed for machine-printed text. In this paper, we present a simple and robust text-line extraction algorithm for printed Arabic text. The presented method is divided into two stages: preprocessing and text-line extraction. It extracts text-lines efficiently, even in small font sizes, by utilizing baselines, projection profiles, and a top-down divide and conquer technique. The presented method is fast and efficient when dealing with non-uniform inter-line spacing and the text-line overlapping problem.A set of experiments were conducted on the collected dataset. The experiments revealed that the proposed method outperforms two baseline approaches, with an average error rate of 3% on Arabic text without diacritics and 11% on Arabic text with diacritics. Furthermore, the experiments demonstrate that the proposed algorithm has a simple computational running time, with an average running time of 0.087 s per document image.