Deep Learning Neural Network Algorithm Research Articles

Accessing versatile tensile ductility of amorphous materials by fractal nanoarchitecture design

Amorphous materials have been known to be intrinsically brittle under tensile stress. This is especially true in silicon-based and metallic-based glasses. In particular, for the latter, they quite often show promising mechanical properties superior to their crystalline counterparts. However, lacking tensile ductility strongly prohibits them from servicing the societies. Although with decades of immense research efforts, we are still waiting for a sophisticated solution. To march in this direction, in this work, we are dedicated to finding a practical method by smart fractal nano-architecture design through advanced computational modeling. This mainly aims to circumvent the intrinsic strain localization at the nano-scale to avoid catastrophic failure. By distributing the external strain to numerous self-confined nano-branches, we successfully achieve astonishing and tunable tensile ductility. This strategy proves to be very effective for different classes of amorphous materials. We found that the tensile properties of the nano-architectured glasses depend on both the constituent elements and the nanostructure. This demonstrates our flexible capability to design desired mechanical properties for a specific material at any spatial length scale. The current findings will inspire experimental realization by cutting-edge 3D-printing techniques and call for optimal design by top-notch graph neural network deep learning algorithms. This work also proposes a new ’structure’-property relationship to efficiently bridge experimental fabrication and computational design.

Digital versus manual pathology scoring of PD-L1 in patients with esophageal squamous cell carcinoma (ESCC).

e16023 Background: Immune checkpoint inhibitors (ICIs) are standard-of-care for patients with ESCC, but the association between PD-L1 and clinical outcomes of patients with ESCC receiving ICIs treatment is unclear. Currently, different PD-L1 antibody clones, scoring, and cutoffs were used for different ICIs development in ESCC. Thus, we develop a digital pathology (DP) algorithm to generate multiple scores at the same time with high reproducibility. We evaluate DP algorithm by: (1) comparing the digital and manual scores of SP263 and 22C3 assays; (2) evaluating the association with OS. Methods: Tumor tissue samples were collected from ESCC patients treated at National Taiwan University Hospital between 2017 and 2020. Slides were tested for PD-L1 using SP263 and 22C3 assays at a College of American Pathologists accredited central laboratory. H&E and PD-L1 slides were scanned for whole slide images (WSIs) using DP200 scanner at 20x. HALO (v3.5) by indica labs was used for DP development by deep learning neural network algorithms. WSIs of SP263 were annotated for region of interest (ROI), tumor/stroma area, and tumor, immune and fibroblast cells by board certified pathologists. After the algorithm is developed and run on SP263 WSIs, a board-certified pathologist reviewed the images and provided an agreement score of DP markups. The images were excluded if an agreement was less than 80%. This algorithm was then applied to 22C3 WSIs. Data of areas and cells were exported for subsequent statistical analysis using R. We calculated the DP scores based on the definition of manual readouts of TAP, CPS and s/iTILs in ROI and tumor and stroma regions in both clones’ WSIs. These DP scores were compared with manual scores using Spearman’s rank-order correlation. Association with OS were evaluated using cox proportional hazard regression. Results: Of 235 evaluable images, twelve (5.1%) were excluded due to the low agreement with the pathologist's judgment. DP generated scores showed a strong correlation with pathologists scoring (rho = 0.80 for SP263 and rho = 0.74 for 22C3). DP generated CPS-like and TAP-like scores in ROI, and tumor and stroma regions from SP263 and 22C3 also demonstrated a strong correlation between each other (rho: 0.77 - 0.86). However, s/iTILs correlation between DP and pathologist score is negligible (rho: 0.002 - 0.34). While examining the association of PD-L1 scores and s/iTILs with OS, no clear associations were found with either DP or pathologists’ scores, after adjusting for stage, ECOG, age and gender. Conclusions: DP algorithms developed in this study demonstrated the efficiency of generating multiple DP scores of CPS, TAP, and s/iTILs from one image at the same time, which performed comparable to manual scores. DP algorithm developed based on SP263 is generalizable to 22C3 and showed a good performance. Further algorithm adjustment will be made to overcome the misclassification in the 12 failed cases.

EnDL-HemoLyt: Ensemble Deep Learning-based Tool for Identifying Therapeutic Peptides with Low Hemolytic Activity.

Low hemolytic therapeutic peptides have gained an edge over small molecule-based medicines. However, finding low hemolytic peptides in laboratory is time-consuming, costly and necessitates the use of mammalian red blood cells. Therefore, wet-lab researchers often perform in-silico prediction to select low hemolytic peptides before proceeding with in-vitro testing. The in-silico tools available for this purpose have following limitations: (i) They do not provide predictions for peptides having N/C terminal modifications. (ii) Data is food for AI; however, datasets used to create existing tools do not contain peptide data generated over past eight years. (iii) Performance of available tools is also low. Therefore, a novel framework has been proposed in current work. Proposed framework utilizes recent dataset and uses ensemble learning technique to combine the decisions produced by bidirectional long short-term memory, bidirectional temporal convolutional network, and 1-dimensional convolutional neural network deep learning algorithms. Deep learning algorithms are capable of extracting features themselves from data. However, instead of relying solely on deep learning-based features (DLF), handcrafted features (HCF) were also provided so that deep learning algorithms can learn features that are missing from HCF, and a better feature vector can be constructed by concatenating HCF and DLF. Additionally, ablation studies were carried out to understand the roles of an ensemble algorithm, HCF, and DLF in the proposed framework. Ablation studies found that the ensemble algorithm, HCF and DLF are crucial components of proposed framework, and there is a decrease in performance on eliminating any of them. Mean value of performance metrics, namely Acc, Sn, Pr, Fs, Sp, Ba, and Mcc obtained by proposed framework for test data is ≈ 87, 85, 86, 86, 88, 87, and 73, respectively. To aid scientific community, model developed from proposed framework has been deployed as a web server at https://endl-hemolyt.anvil.app/.

Outdoor regenerative water purification and drinking water station based on semiconductor condensation technology

In some exhibition sites and outdoor sports venues, there is still a lack of public drinking water resources and hand washing facilities., This topic in view of the existing problems put forward a new scheme, and designed a based on semiconductor condensation technology of outdoor regeneration net drink integration intelligent water station, photovoltaic energy storage technology and air water, realize the self-sufficiency of energy and water, and break the equipment in the regional limitations, make it to meet the needs of people hand clean and healthy drinking water, solve the problem of lack of clean water in specific places. In addition, the device integrates the neural network deep learning algorithm to predict the water consumption, and provides emergency charging and advertising and other functions, which has good social benefits and market prospects.

Automatic fabrication system of optical micro-nanofiber based on deep learning

<sec>The wide range, high precision, and dynamic real-time measurement of micro-/nanofiber diameter are crucial for achieving low loss transmission and controlling dispersion in the preparation process of micro-nanofiber. In view of the problems of small diameter regulation range, complex operation and long-time consumption of the existing preparation methods, the automatic detection system of micro-nanofiber is realized based on deep learning neural network algorithm in this work. The image segmentation method in computer vision is used to make high-quality multi-scale micro-/nanofiber datasets, and the improved YOLOv8-FD (You Only Look Once version 8-Fiber Detection) algorithm based on small target detection is used to automatically detect the diameter of micro-nanofiber.</sec><sec>Through image segmentation and identification of the target of single pixel size in the microscopic image, the diameter detection of micro-nanofiber is finally realized. In this process, the real-time diameter of micro-nanofiber is obtained through image information, and then the micro-nanofiber small target is accurately segmented to achieve the precise detection of mAP<sup>IoU</sup><sup>=50</sup> = 0.975 and mAP<sup>IoU</sup><sup>=50—95</sup> = 0.765 on the micro-nanofiber multi-scale target dataset with extremely high accuracy. The algorithm-based construction of a high-precision micro-nanofiber automatic preparation system enables real-time accurate segmentation of fiber edges, calculation of fiber diameter, and feedback to the control system for achieving automated preparation of fibers with arbitrary diameters. Additionally, it facilitates the detection of micro-nanofiber in a range from 462 nm to 125 μm. The average response time for reasoning is 9.6 ms, and the detection error is kept below 2.95%.</sec><sec>In addition, compared with other micro-/nanofiber diameter detection methods based on optical imaging and mode cutoff, this method shows advantages of high precision, high speed and arbitrary diameter preparation for diameter detection based on deep learning neural networks. The system is very suitable for high-precision real-time measurement and automatic and accurate preparation of micro-/nanofibers, thereby providing a novel method of developing micro-nanofiber devices with low-loss transmission and adjustable dispersion. </sec>

Image processing and machine learning for diagnosis and screening of craniosynostosis in children

Objectivecraniosynostosis (CSO) is a congenital disorder resulting from early closure of cranial sutures in newborns, while could cause significant cosmetic and neurodevelopmental problems. As a standard method, different craniometric indices are measured directly from child head or from their 3D CT scan of skull for diagnosis or in post-operative follow-up period. We propose a novel telehealth-compatible deep learning neural network-based method for identifying different craniometric indices in non-syndromic CSO patients 2D photographic data. Methods624 pre-operative and post-operative top-down cranial digital images of 145 craniosynostotic infants (59 sagittal, 55 metopic and 31 unicoronal synostosis) who had surgery at Mofid Children’s Hospital, Tehran, Iran were used in a deep learning neural network algorithm. Head boundary was defined by a faster region-based convolutional neural network (Faster R-CNN) and then different cranial indices (cranial index (CI), cranial vault asymmetry index (CVAI), anterior-posterior width ratio (APWR), anterior-midline width ratio (AMWR) and left–right height ratio (LRHR)) were calculated from segmented images. Accuracy, sensitivity and specificity were calculated for software versus specialist data association between cranial indices were evaluated with inter-class correlation coefficients. ResultsThe head border was segmented in the proposed images with accuracy of 88.67 ± 1.94 in comparison with standard hand made procedure with a sensitivity of 86.91 ± 3.75 and specificity of 88.60 ± 4.81. Among calculated cranial indices, significant decrease in CI value is most useful for diagnosis of sagittal synostosis (CIsagittal = 71.97 ± 4.33), significant increase in CVAI value and significant decrease in LRHR value is most appropriate for unicoronal suture synostosis diagnosis (CVAIunicoronal=6.79±3.80 and LRHRunicoronal = 0.91 ± 0.05) and significant decrease in APWR and AMWR values could be indicator of metopic synostosis (AMWRmetopic= 0.77 ± 0.04 and APWRmatopic = 0.83 ± 0.05). ConclusionDeep learning neural network algorithms could have high levels of capability in calculating cranial indices from routine 2D digital images of non-syndromic craniosynostotic children and act as a substitute for optical scanner or 3D CT-based craniometrics. This method could act as a corner stone for developing a software for a mobile platform that that would allow for screening by tele-medicine or in a primary care setting.

A Study of Deep Learning Neural Network Algorithms and Genetic Algorithms for FJSP

Flexible job-shop scheduling problem (FJSP) is a new research hotspot in the field of production scheduling. To solve the multiobjective FJSP problem, the production of flexible job shop can run normally and quickly. This research takes into account various characteristics of FJSP problems, such as the need to ensure the continuity and stability of processing, the existence of multiple objectives in the whole process, and the constant complexity of changes. It starts with deep learning neural networks and genetic algorithms. Long short-term memory (LSTM) and convolutional neural networks (CNN) are combined in deep learning neural networks. The new improved algorithm is based on the combination of deep learning neural networks LSTM and CNN with genetic algorithm (GA), namely, CNN-LSTM-GA algorithm. Simulation results showed that the accuracy of the CNN-LSTM-GA algorithm was between 85.2% and 95.3% in the test set. In the verification set, the minimum accuracy of the CNN-LSTM-GA algorithm was 84.6%, both of which were higher than the maximum accuracy of the other two algorithms. In the FJSP simulation experiment, the AUC value of the CNN-LSTM-GA algorithm was 0.92. After 40 iterations, the F1 value of the CNN-LSTM-GA algorithm remained above 0.8, which was significantly higher than the other two algorithms. CNN-LSTM-GA is superior to the other two algorithms in terms of prediction accuracy and overall performance of FJSP. It is more suitable for solving the discrete manufacturing job scheduling problem with FJSP characteristics. This study significantly raises the utilisation rate of the assembly shop’s equipment, optimises the scheduling of FJSP, and fully utilises each processing device’s versatile characteristics, which are quite useful for the production processes of domestic vehicle manufacturing companies.

Deep learning for COVID‐19 contamination analysis and prediction using ECG images on Raspberry Pi 4

AbstractThis paper's primary goal is to diagnose COVID‐19 contamination based on the artificial intelligence approach automatically. We used convolutional neural network deep learning algorithm for analyzing the ECG images to detect cardiac abnormalities, consequent of the contamination by the SARS‐CoV‐2 virus, responsible for the COVID‐19 epidemic. We designed, trained, and evaluated the performance of two deep learning models (MobileNetV2 and VGG16) in detecting and distinguishing between two different classes (healthy subjects and COVID‐19 positive cases). Indeed, this virus attacks the human respiratory system, which could affect the heart system. Thus, developing a deep learning model could help for a quick and efficient diagnosis, prediction, and physician decision‐making. The performed deep learning model will be used for predicting abnormal cardiac activities consequent to the contamination by the virus. The overall classification rate achieved by the models was 99.34% and 99.67% for MobileNetV2 and VGG16, respectively. Therefore, this approach can efficiently contribute to the diagnosis of COVID‐19 contamination.

Prediction of Bearing Vibration Fault State based on Fused Bi-LSTM and SVM

Machinery failure prediction and equipment health management is the key to reduce even prevent the occurrence of significant failures. In recent years, deep learning based intelligent fault diagnosis methods have become a hot topic of research in the field of machinery and equipment health management, especially the evaluation of the equipment health indicators and failure mode identification issues in complex environments. In this paper, a fused bidirectional long and short term memory (BiLSTM) and support vector machine (SVM) recurrent neural network deep learning algorithm is proposed to solve the problem of fast fault diagnosis of mechanical bearing vibration in complex environments. The algorithm uses a BiLSTM memory network as a feature extractor to capture fault features in the raw one dimensional (1-D) time series data, and then a SVM is used at the end of the network instead of the traditional softmax function as a classifier discriminator to further improve the accuracy of diagnosis. The experimental results indicate that the proposed algorithm model has higher diagnostic accuracy and faster diagnosis speed, and more suitable for fast diagnosis and real time detection of faults in complex environments.

A stress defect state measurement method based on low-frequency ACMFL excitation and Hall sensor array collection

The safety testing of ferromagnetic materials, which are the main materials for various machines and equipment, is particularly important. Stress concentration zones (stress defects) cause stress corrosion of ferromagnetic materials, and also have the potential to cause direct damage to ferromagnetic materials. Estimation of stress sources state using electromagnetic nondestructive measurement methods is a critical and difficult problem. In this paper, a visual and intelligent identification method of stress defects in ferromagnetic materials by low frequency AC magnetic flux leakage (ACMFL) technique is proposed. A new three-point compression experiment was designed in this paper. Time-difference vision is established to analyze the ACMFL signal caused by stress defects. A visual transformed convolutional neural network deep learning algorithm has been proposed to identify grayscale patterns pre-processed by the time-difference vision. The results show that the method proposed in this paper elucidates the relationship between the time-difference vision of a stress defect and the stress source state of the mechanical stress. Our proposed method allows to analyze the pressure indenter size of the pressure source of stress defects.

Deep Learning–Based Estimation of Implantable Collamer Lens Vault Using Optical Coherence Tomography

To develop and validate a deep learning neural network for automated measurement of implantable collamer lens (ICL) vault using anterior segment optical coherence tomography (AS-OCT). Cross-sectional retrospective study. Multicentric. A total of 2,647 AS-OCT anonymized eye scans were used from 139 eyes of 82 subjects who underwent ICL surgery in 3 different centers. Using transfer learning, a deep learning neural network algorithm was trained and validated for the regression task of estimating the ICL vault on the MS-39 OCT (CSO, Florence, Italy). A trained operator separately reviewed all OCT scans and measured the central vault using a built-in OCT caliper tool. The model was then separately tested on 191 AS-OCT scans. A Bland-Altman plot was constructed and the mean absolute percentage error (MAPE), the mean absolute error (MAE), the root mean squared error (RMSE), the Pearson correlation coefficient (r), and the determination coefficient (R2) were calculated to evaluate the strength and validity of the model. On the test set, the model achieved a MAPE of 3.42%, an MAE of 15.82 µm, an RMSE of 18.85 µm, a Pearson correlation coefficient r of +0.98 (P-Value < .00001), and a coefficient of determination R2 of +0.96. There was no significant difference between the vaults of the test set labeled by the technician versus those estimated by the model: 478 ± 95 µm versus 475 ± 97 µm, respectively, P-Value = 0.064). Using transfer learning, our deep learning neural network was able to accurately compute the ICL vault from AS-OCT scans, overcoming the limitations of an imbalanced dataset and limited training data. Such an algorithm can assist the postoperative assessment in ICL surgery. PRéCIS: This study develops and validates a deep learning model to accurately estimate implantable collamer lens vault on optical coherence tomography using transfer learning, convoluted neural networks, and regression.

Classification of Ayurveda constitution types: a deep learning approach

In precision medicine, there has been a paradigm shift, from a population average based one-size-fits all, to an individualized approach. The aim is to tailor a medicine according to an individual’s genetic make-up and response to drug, diet, and environment. Machine learning algorithms are being developed to dissect inter-individual variability in the population. This is more important for prevention of diseases in healthy individuals and early actionable interventions. In Ayurveda, the basic tenet of the practice is personalized—an individual’s constitution type “Prakriti” is used for predicting healthy and disease trajectories as well as individualized management. Prakriti refers to the nature of the body in terms of doshas, and the time of conception is decided according to the predominance of doshas. It does not change throughout life and is responsible for a person’s physical and mental characteristics. Any population can be stratified into seven Prakriti groups amongst which the Vata, Pitta and Kapha are the three phenotypic extremes with contrasting disease susceptibilities and response. Prakriti is assessed through examination of multi-system attributes and has been formalized into a questionnaire that captures 120 feature attributes for objective assessment. We use for the first time—powerful dense neural network deep learning algorithms for predicting Prakriti classes. The study was carried out on a validated dataset of 233 extreme Prakriti individuals derived from two genetically diverse cohorts from North and West India. We benchmark our method with the state-of-the-art supervised machine learning and establish a learning approach for transferring features from the Western to Northern Indian data and vice versa as well. Our study provides a computational framework for predicting Prakriti which can be implemented in large scale studies of precision medicine. The dataset and the baseline implementation can be found in ( https://github.com/skarifahmed/DeepAyurveda )

Online course resource recommendation based on deep learning

In order to further improve the online learning experience and effect of courses for learners, this study, with the help of deep learning neural network algorithm, based on the characteristics of deep learning algorithm such as good processing performance and strong data adaptability, carries out research for online course students' preferences, learner characteristics and other fields, and constructs a course resource recommendation technical algorithm based on deep learning. Through algorithm experiments, it is found that the performance of NCF model is improved by 57.7% compared with that of CF model in the algorithm with RMSE as evaluation index. The performance under MAE evaluation index is improved by 58.5%, indicating that the integration of the deep learning neural network algorithm can better improve the online course recommendation effect and enhance students' online learning experience.

Using Deep Learning Algorithms to Diagnose Coronavirus Disease (COVID-19)

With the rapid development in the area of Machine Learning (ML) and Deep learning, it is important to exploit these tools to contribute to mitigating the effects of the coronavirus pandemic. Early diagnosis of the presence of this virus in the human body can be crucially helpful to healthcare professionals. In this paper, three well-known Convolutional Neural Network deep learning algorithms (VGGNet 16, GoogleNet and ResNet50) are applied to measure their ability to distinguish COVID-19 patients from other patients and to evaluate the best performance among these algorithms with a large dataset. Two stages are conducted, the first stage with 14994 x-ray images and the second one with 33178. Each model has been applied with different batch sizes 16, 32 and 64 in each stage to measure the impact of data size and batch size factors on the accuracy results. The second stage achieved accuracy better than the first one and the 64 batch size gain best results than the 16 and 32. ResNet50 achieves a high rate of 99.31, GoogleNet model achieves 95.55, while VGG16 achieves 96.5. Ultimately, the results affect the process of expediting the diagnosis and referral of these treatable conditions, thereby facilitating earlier treatment, and resulting in improved clinical outcomes.

Comparative Analysis of Deep Learning Models on Brain Tumor Segmentation Datasets: BraTS 2015-2020 Datasets

Deep Learning neural networks have shown applicability in segmentation of brain tumor images.This research have been carried for comprehensive review of several deep learning neural networks. The datasets included in this study are standard datasets Multimodal Brain Tumor Segmentation (BraTS). This paper has summarized the performance of various deep learning neural network algorithms on BraTS datasets. Algorithms have been compared and summarized against the baseline models with specific attributes like dice score, PPV and sensitivity. It has been found that out of the different models applied on the BraTS 2015 dataset GAN in the year 2020 algorithm is showing better results on this data set. GAN architecture termed RescueNet gave the best segmentation results in terms of 0.94 dice score and 0.88 Sensitivity. This has been also observed that models used cascaded deep learning models had independent deep learning models at each stage which had no correlation among the stages which can cause class imbalance. Further it have found that the Attention models tried to solve problem of class imbalance in the brain tumor segmentation task. This work also found that existing CNN’s is having overfitting issues. For this ResNet models can add a rapid connect bounce relationship parallel to the layers of CNN to accomplish better outcomes for the brain tumor segmentation task.

An Enhanced LBPH Approach to Ambient-Light-Affected Face Recognition Data in Sensor Network

Although combining a high-resolution camera with a wireless sensing network is effective for interpreting different signals for image presentation on the identification of face recognition, its accuracy is still severely restricted. Removing the unfavorable impact of ambient light remains one of the most difficult challenges in facial recognition. Therefore, it is important to find an algorithm that can capture the major features of the object when there are ambient light changes. In this study, face recognition is used as an example of image recognition to analyze the differences between Local Binary Patterns Histograms (LBPH) and OpenFace deep learning neural network algorithms and compare the accuracy and error rates of face recognition in different environmental lighting. According to the prediction results of 13 images based on grouping statistics, the accuracy rate of face recognition of LBPH is higher than that of OpenFace in scenes with changes in ambient lighting. When the azimuth angle of the light source is more than +/−25° and the elevation angle is +000°, the accuracy rate of face recognition is low. When the azimuth angle is between +25° and −25° and the elevation angle is +000°, the accuracy rate of face recognition is higher. Through the experimental design, the results show that, concerning the uncertainty of illumination angles of lighting source, the LBPH algorithm has a higher accuracy in face recognition.

Pattern Recognition of Wushu Routine Action Decomposition Process Based on Kinect

Human action recognition is a hotspot in the fields of computer vision and pattern recognition. Human action recognition technology has created huge social value and considerable economic value for the society. Meeting people’s needs and understanding people’s expressions are the current research focus. Aiming at the problem that the movement cannot be continuously identified and due to a lack of detailed features in the action decomposition pattern recognition in the traditional Wushu routine decomposition process, it is proposed to use Kinect technology to identify the Wushu routine movement decomposition process in the Wushu routine movement decomposition process. This paper analyzes the principle of skeleton tracking and skeleton extraction performed by the Kinect human sensor and uses the Kinect sensor with the Visual Studio 2015 development platform to collect and process the skeleton data of limb movements and defines eight static limb motion samples and four dynamic limbs. The study uses a deep learning neural network algorithm to train and identify the established database of static body movements and uses the same template matching algorithm and K-NN. The recognition effects of the algorithms were compared and analyzed, and it was concluded that the static body motion recognition rates of the three algorithms were all above 90%. In this paper, recognition experiments are carried out on the MSR action 3D database. The influence of different integrated decision-making methods on the recognition results is further discussed and analyzed, and the average method integrated decision-making, which is most suitable for the algorithm model in this paper, is proposed. The results show that the recognition accuracy of the algorithm reaches 98.1%, which proves the feasibility of the preprocessing algorithm.

Functional mimicry of Ruffini receptors with fibre Bragg gratings and deep neural networks enables a bio-inspired large-area tactile-sensitive skin

Collaborative robots are expected to physically interact with humans in daily living and the workplace, including industrial and healthcare settings. A key related enabling technology is tactile sensing, which currently requires addressing the outstanding scientific challenge to simultaneously detect contact location and intensity by means of soft conformable artificial skins adapting over large areas to the complex curved geometries of robot embodiments. In this work, the development of a large-area sensitive soft skin with a curved geometry is presented, allowing for robot total-body coverage through modular patches. The biomimetic skin consists of a soft polymeric matrix, resembling a human forearm, embedded with photonic fibre Bragg grating transducers, which partially mimics Ruffini mechanoreceptor functionality with diffuse, overlapping receptive fields. A convolutional neural network deep learning algorithm and a multigrid neuron integration process were implemented to decode the fibre Bragg grating sensor outputs for inference of contact force magnitude and localization through the skin surface. Results of 35 mN (interquartile range 56 mN) and 3.2 mm (interquartile range 2.3 mm) median errors were achieved for force and localization predictions, respectively. Demonstrations with an anthropomorphic arm pave the way towards artificial intelligence based integrated skins enabling safe human–robot cooperation via machine intelligence.

Characteristics, Main Controlling Factors and Prediction of Complex Fracture–Cavity Reservoirs of Buried Carbonate Hills in the Weixinan Depression, Western South China Sea

Buried hill oil reservoirs have become a key area for offshore oil and gas exploration. In this paper, a typical oil field in the western South China Sea is used as the research object, and a study on the characterization, cause of formation and prediction of the fracture–cavity reservoir distribution is carried out. The reservoir in the study area is a complex fracture–cavity reservoir that developed due to weathering and leaching, tectonic movement and dissolution reconstruction on the limestone skeleton. The reservoir spaces are composed of karst caves, fractures and pores. The main controlling factors include lithological changes, karst landforms, tectonic deformation and faulting. To address the controlling mechanisms of the lithological changes on the formation of fracture–cavity reservoirs, a new parameter, the lithology standard deviation, to evaluate lithological changes is proposed based on the characteristics of the lithological changes, and the distribution of these lithological changes is portrayed in combination with the seismic attributes. The tectonic deformation principal curvature inversion algorithm is used to simulate the distribution of the tectonic principal curvature at the top of the Carboniferous. The larger the tectonic principal curvature is, the stronger the deformation of the rock formation and the more favorable the conditions are for fracture formation. The karst geomorphology controls the overall reservoir distribution, and the karst highlands and karst slope areas are the zones with the most–developed secondary pore space (or fractures and karst caves). The faulting control area is the fracture and dissolution pore development area, major faults control the distribution of the karst cave reservoirs, and secondary faults influence the formation of fractures in the faulted area. The study predicts and evaluates the distribution of fracture–cavity reservoirs from the perspective of fracture–cavity genesis quantification, and by gridding and normalizing the four major genesis quantification evaluation parameters and fusing the geological factors that control the formation of fractures and karst caves by using Back–Propagation neural network deep learning algorithms, a method for predicting the distribution of fracture–cavity reservoirs constrained by geological genesis analysis is developed.

Evaluation of market risk and resource allocation ability of green credit business by deep learning under internet of things.

The research expects to evaluate the capital market risk and resource allocation ability of green credit business exploration based on neural network algorithm by deep learning in the context of the Internet of things, increase the funds flowing to green environmental protection industry, accelerate the development of real economy and stabilize China’s market economy. On the basis of previous studies, the research takes the credit business in the capital market as the research object, and improves the ability of resource allocation by optimizing the financial transaction structure. On this basis, through comparative analysis, the grey system model is implemented. back propagation neural network model under deep learning is used to evaluate the capital market risk of green credit business exploration, and the data of different provinces in China from 2009 to 2019 are taken as an example to verify. The model is used to measure the relationship between green credit business and industrial structure. Additionally, it also analyzes the main factors affecting the efficiency of green credit. The results show that green credit mainly affects the industrial structure through enterprise capital and financing channels. China’s overall green credit adjustment has had a significant upgrading effect on the industrial structure. The impact of green credit on industrial structure adjustment is different in the east, middle, and west regions. Optimizing the project capital structure, promoting seasonal financial transformation, setting up the function of innovation platform, and improving the internal governance structure of enterprises can improve financing efficiency and realize green and sustainable economic development in the future. The research results can provide a theoretical basis for the green development of China’s financial market and the application of deep learning neural network algorithm under the background of Internet of things.