Search

Gear and its transmission are widely used in different transmission systems, and its complicated and changeable condition brings a series of problems to the fault feature extraction and diagnosis. In recent years, deep learning techniques have been gradually applied to feature extraction and pattern recognition, and the features of feature extraction and fault diagnosis in complex working environments have shown certain advantages. This study is based on stacked autoencoder under deep learning model, and improve training network performance by modified activation function. Through the network training before and after the experiment done, and to extract the fault feature data comparison in testing, improving network after activation function to extract fault features showed a greater advantage, can be a very good application in practical fault feature extraction.

The SARS-CoV-2 virus is responsible for the rapid global spread of the COVID-19 disease. As a result, it is critical to understand and collect primary data on the virus, infection epidemiology, and treatment. Despite the speed with which the virus was detected, studies of its cell biology and architecture at the ultrastructural level are still in their infancy. Therefore, we investigated and analyzed the viral morphometry of SARS-CoV-2 to extract important key points of the virus's characteristics. Then, we proposed a prediction model to identify the real virus levels based on the optimization of a full recurrent neural network (RNN) using transmission electron microscopy (TEM) images. Consequently, identification of virus levels depends on the size of the morphometry of the area (width, height, circularity, roundness, aspect ratio, and solidity). The results of our model were an error score of training network performance 3.216 × 10−11 at 639 epoch, regression of −1.6 × 10−9, momentum gain (Mu) 1 × 10−9, and gradient value of 9.6852 × 10−8, which represent a network with a high ability to predict virus levels. The fully automated system enables virologists to take a high-accuracy approach to virus diagnosis, prevention of mutations, and life cycle and improvement of diagnostic reagents and drugs, adding a point of view to the advancement of medical virology. [ABSTRACT FROM AUTHOR]

Simple Summary: Inflammatory bowel disease includes ulcerative colitis and Crohn's disease. Ulcerative colitis affects the colon; its pathogenesis involves genetic susceptibility, microbes, and immune dysregulation, and a higher risk of colorectal cancer. This study classified images of ulcerative colitis using deep learning. A dataset was created to process images of the large intestine capturing the three diagnoses of ulcerative colitis, colorectal cancer (adenocarcinoma), and normal colon. The convolutional neural network (CNN) was trained to classify the images into three diagnostic classes, and the performance was tested on an independent dataset. The gradient-weighted class activation mapping (Grad-CAM) heatmap technique was used to understand the classification decisions. Finally, LAIR1 and TOX2 expressions were analyzed in the ulcerative colitis cases. In conclusion, the network classified the three diagnoses with high performance, and LAIR1 and TOX2 were found to correlate with the severity of ulcerative colitis. Background: Ulcerative colitis is a chronic inflammatory bowel disease of the colon mucosa associated with a higher risk of colorectal cancer. Objective: This study classified hematoxylin and eosin (H&E) histological images of ulcerative colitis, normal colon, and colorectal cancer using artificial intelligence (deep learning). Methods: A convolutional neural network (CNN) was designed and trained to classify the three types of diagnosis, including 35 cases of ulcerative colitis (n = 9281 patches), 21 colon control (n = 12,246), and 18 colorectal cancer (n = 63,725). The data were partitioned into training (70%) and validation sets (10%) for training the network, and a test set (20%) to test the performance on the new data. The CNNs included transfer learning from ResNet-18, and a comparison with other CNN models was performed. Explainable artificial intelligence for computer vision was used with the Grad-CAM technique, and additional LAIR1 and TOX2 immunohistochemistry was performed in ulcerative colitis to analyze the immune microenvironment. Results: Conventional clinicopathological analysis showed that steroid-requiring ulcerative colitis was characterized by higher endoscopic Baron and histologic Geboes scores and LAIR1 expression in the lamina propria, but lower TOX2 expression in isolated lymphoid follicles (all p values < 0.05) compared to mesalazine-responsive ulcerative colitis. The CNN classification accuracy was 99.1% for ulcerative colitis, 99.8% for colorectal cancer, and 99.1% for colon control. The Grad-CAM heatmap confirmed which regions of the images were the most important. The CNNs also differentiated between steroid-requiring and mesalazine-responsive ulcerative colitis based on H&E, LAIR1, and TOX2 staining. Additional classification of 10 new cases of colorectal cancer (adenocarcinoma) were correctly classified. Conclusions: CNNs are especially suited for image classification in conditions such as ulcerative colitis and colorectal cancer; LAIR1 and TOX2 are relevant immuno-oncology markers in ulcerative colitis. [ABSTRACT FROM AUTHOR]

Discovering fake news is very important in the press to maintain credibility and combat wrong information. Traditional fake news detection systems (FNDS) are challenged due to accuracy problems due to the complex nature of misinformation in news articles. This study introduces a new approach using a group method that combines automatic learning algorithms such as naive Bays, K-NN, Random Forest, Decision Tree, and Support Vector Machine (SVM) by integrating majority votes. The results indicate that group technologies greatly enhance the accuracy of defining the fake news compared to the algorithm-based works. By leveraging collective intelligence, FNDS addresses the complexities of fake news detection. This approach enhances its capability to identify and refute misinformation effectively. This research emphasizes the choice of appropriate algorithms and the integration of the group's methods to develop the most accurate and strong fake news discovery systems, contributing to combating wrong information in the press. The proposed approach achieved an accuracy improvement of 15% compared to individual algorithms, with an overall F1-score of 89.5%. [ABSTRACT FROM AUTHOR]

Gear and its transmission are widely used in different transmission systems, and its complicated and changeable condition brings a series of problems to the fault feature extraction and diagnosis. In recent years, deep learning techniques have been gradually applied to feature extraction and pattern recognition, and the features of feature extraction and fault diagnosis in complex working environments have shown certain advantages. This study is based on stacked autoencoder under deep learning model, and improve training network performance by modified activation function. Through the network training before and after the experiment done, and to extract the fault feature data comparison in testing, improving network after activation function to extract fault features showed a greater advantage, can be a very good application in practical fault feature extraction. [ABSTRACT FROM AUTHOR]

Accurate biomass estimates are key to understanding a wide variety of ecological functions. In marine systems, epibenthic biomass estimates have traditionally relied on either destructive/extractive methods that are limited to horizontal soft‐sediment environments, or simplistic geometry‐based biomass conversions that are unsuitable for more complex morphologies. Consequently, there is a requirement for non‐destructive, higher‐accuracy methods that can be used in an array of environments, targeting more morphologically diverse taxa, and at ecological relevant scales.We used a combination of 3D photogrammetry, convolutional neural network (CNN) automated taxonomic identification, and taxa‐specific biovolume:biomass calibrations to test the viability of estimating biomass of three species of morphologically complex epibenthic taxa from in situ stereo 2D source imagery.Our trained CNN produced accurate and reliable annotations of our target taxa across a wide range of conditions. When incorporated into photogrammetric 3D models of underwater surveys, we were able to automatically isolate our three target taxa from their environment, producing biovolume measurements that had respective mean similarities of 99%, 102% and 120% of those obtained from human annotators. When combined with taxa‐specific biovolume:biomass calibration values, we produced biomass estimates of 88%, 125% and 133% mean similarity to that of the 'true' biomass of the respective taxa.Our methodology provides a highly reliable and efficient method for estimating epibenthic biomass of morphologically complex taxa using non‐destructive 2D imagery. This approach can be applied to a variety of environments and photo/video survey approaches (e.g. SCUBA, ROV, AUV) and is especially valuable in spatially extensive surveys where manual approaches are prohibitively time‐consuming. [ABSTRACT FROM AUTHOR]

Ceramic tiles are one of themost indispensablematerials for interior decoration. The ceramic patterns can't match the design requirements in terms of diversity and interactivity due to their natural textures. In this paper, we propose a sketch-based generation method for generating diverse ceramic tile images based on a hand-drawn sketches using Generative Adversarial Network (GAN). The generated tile images can be tailored tomeet the specific needs of the user for the tile textures. The proposed method consists of four steps. Firstly, a dataset of ceramic tile images with diverse distributions is created and then pre-trained based on GAN. Secondly, for each ceramic tile image in the dataset, the corresponding sketch image is generated and then the mapping relationship between the images is trained based on a sketch extraction network using ResNet Block and jump connection to improve the quality of the generated sketches. Thirdly, the sketch style is redefined according to the characteristics of the ceramic tile images and then double cross-domain adversarial loss functions are employed to guide the ceramic tile generation network for fitting in the direction of the sketch style and to improve the training speed. Finally, we apply hidden space perturbation and interpolation for further enriching the output textures style and satisfying the concept of "one style with multiple faces". We conduct the training process of the proposed generation network on 2583 ceramic tile images dataset. To measure the generative diversity and quality, we use Frechet Inception Distance (FID) and Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE) metrics. The experimental results prove that the proposed model greatly enhances the generation results of the ceramic tile images, with FID of 32.47 and BRISQUE of 28.44. [ABSTRACT FROM AUTHOR]

Earlier studies focused on training ResNet50 deep learning models on a dataset of fundus images from the National Taiwan University Hospital HsinChu Branch. The study aimed to identify class-specific discriminative areas related to various conditions of ganglion cell complex (GCC) thickness, center focus areas, cropped patches from the fundus, and dataset partitions. The study utilized two visualization methods to evaluate and explain the areas of interest of the network model and determine if they aligned with clinical diagnostic knowledge. The results of the experiments demonstrated that incorporating GCC thickness information improved the accuracy of glaucoma determination. The deep learning models primarily focused on the optic nerve head (ONH) for glaucoma diagnosis, which was consistent with clinical rules. Nonetheless, the models achieved high prediction accuracy in detecting glaucomatous cases using only cropped images of macular areas. Moreover, the model's focus on regions with GCC impairment in some cases indicates that deep learning models can identify morphologically detailed alterations in fundus photographs that may be beyond the scope of visual diagnosis by experts. This highlights the significant contribution of deep learning models in the diagnosis of glaucoma. [ABSTRACT FROM AUTHOR]

Person re-identification (Re-ID) aims to retrieve a particular pedestrian's identification from a surveillance system consisting of non-overlapping cameras. In recent years, researchers have begun to focus on open-world person Re-ID tasks based on non-ideal situations. One of the most representative of these is cross-modal person Re-ID, which aims to match probe data with target data from different modalities. According to the modalities of probe and target data, we divided cross-modal person Re-ID into visible–infrared, visible–depth, visible–sketch, and visible–text person Re-ID. In cross-modal person Re-ID, the most challenging problem is the modal gap. According to the different methods of narrowing the modal gap, we classified the existing works into picture-based style conversion methods, feature-based modality-invariant embedding mapping methods, and modality-unrelated auxiliary information mining methods. In addition, by generalizing the aforementioned works, we find that although deep-learning-based models perform well, the black-box-like learning process makes these models less interpretable and generalized. Therefore, we attempted to interpret different cross-modal person Re-ID models from a mathematical perspective. Through the above work, we attempt to compensate for the lack of mathematical interpretation of models in previous person Re-ID reviews and hope that our work will bring new inspiration to researchers. [ABSTRACT FROM AUTHOR]

Robust learning is an important issue in Scientific Machine Learning (SciML). There are several works in the literature addressing this topic. However, there is an increasing demand for methods that can simultaneously consider all the different uncertainty components involved in SciML model identification. Hence, this work proposes a comprehensive methodology for uncertainty evaluation of the SciML that also considers several possible sources of uncertainties involved in the identification process. The uncertainties considered in the proposed method are the absence of a theory, causal models, sensitivity to data corruption or imperfection, and computational effort. Therefore, it is possible to provide an overall strategy for uncertainty-aware models in the SciML field. The methodology is validated through a case study developing a soft sensor for a polymerization reactor. The first step is to build the nonlinear model parameter probability distribution (PDF) by Bayesian inference. The second step is to obtain the machine learning model uncertainty by Monte Carlo simulations. In the first step, a PDF with 30,000 samples is built. In the second step, the uncertainty of the machine learning model is evaluated by sampling 10,000 values through Monte Carlo simulation. The results demonstrate that the identified soft sensors are robust to uncertainties, corroborating the consistency of the proposed approach. [ABSTRACT FROM AUTHOR]

With the rapid development of deep learning and computing power, human–computer interactions, and interfaces are attracting attentions in industrial and academic research. Flexible human–computer interaction can greatly improve productivity and enable robots to work in extreme environments that humans cannot tolerate. The research of gesture recognition is emerging and provides a new way of studying the human–computer interactions. However, compared with the entire human body, human hands are dexterous organs with more complex and flexible joints, which makes hand gesture recognition a challenging problem. Here, a robust and cost‐effective gesture recognition system is reported through the soft optoelectronic sensors. An array of polymer‐encapsulated U‐shaped microfiber (UMF) attached to a glove is fabricated for sensitive finger motion detection. The anisotropic strain response of UMF is measured with a sensitivity of 15.98 (2.20) in the x‐direction (y‐direction). A deep learning network (VGGNet) is developed to process the optical signals for analyzing and classifying hand gestures. The experiments show that VGGNet has high recognition accuracy of 99.2% for the test datasets with ten classified gestures. This work provides a potential optical interface in studying gesture recognition and biomechanical signatures, which can also be applied in virtual reality systems and interactive game platforms. [ABSTRACT FROM AUTHOR]

Purpose: Reading spinal CT (Computed Tomography) images is very important in the diagnosis of spondylosis, which is time-consuming and prones to make biases. In this paper, we propose a framework based on Faster-RCNN to improve detection performances of three spinal fracture lesions: cfracture (cervical fracture), tfracture (thoracic fracture) and lfracture (lumbar fracture). Methods: First, we use ResNet50 to replace VGG16 in backbone network in Faster-RCNN to increase depth of training network. Second, we utilize soft-NMS (Non-Maximum Suppression) instead of NMS to avoid missed detection of overlapped lesions. Third, we simplify RPN (Region Proposal Network) to accelerate training speed and reduce missed detection. Finally, we modify the classifier layer in Faster-RCNN and choose appropriate length-width ratio by changing anchor sizes in sliding window, then adopt multi-scale strategy in training to improve efficiency and accuracy. Results: The experimental results show that the proposed scheme has a good performance, mAP (mean average precision) is 90.6%, IOU (Intersection of Union) is 88.5 and detection time is 0.053 second per CT image, which means our proposed method can accurately detect spinal fracture lesions. Conclusion: Our proposed method can provide assistance and scientific references for both doctors and patients in clinically. [ABSTRACT FROM AUTHOR]

To reduce the manpower consumption on box-level annotations, many weakly supervised object detection methods which only require image-level annotations, have been proposed recently. The training process in these methods is formulated into two steps. They firstly train a neural network under weak supervision to generate pseudo ground truths (PGTs). Then, these PGTs are used to train another network under full supervision. Compared with fully supervised methods, the training process in weakly supervised methods becomes more complex and time-consuming. Furthermore, overwhelming negative proposals are involved at the first step. This is neglected by most methods, which makes the training network biased towards to negative proposals and thus degrades the quality of the PGTs, limiting the training network performance at the second step. Online proposal sampling is an intuitive solution to these issues. However, lacking of adequate labeling, a simple online proposal sampling may make the training network stuck into local minima. To solve this problem, we propose an O nline Active P roposal Set G eneration (OPG) algorithm. Our OPG algorithm consists of two parts: Dynamic Proposal Constraint (DPC) and Proposal Partition (PP). DPC is proposed to dynamically determine different proposal sampling strategies according to the current training state. PP is used to score each proposal, part proposals into different sets and generate an active proposal set for the network optimization. Through experiments, our proposed OPG shows consistent and significant improvement on both datasets PASCAL VOC 2007 and 2012, yielding comparable performance to the state-of-the-art results. • Weakly supervised detection methods are affected by redundant negative proposals. • We propose an online proposal set generation algorithm to removes negative proposals. • Our method improves the detection accuracy without adding much computation. • Our method combines the two-step training process into one-step. [ABSTRACT FROM AUTHOR]

Purpose: Magnetic resonance (MR) imaging is an essential diagnostic tool in clinical medicine. Recently, a variety of deep‐learning methods have been applied to segmentation tasks in medical images, with promising results for computer‐aided diagnosis. For MR images, effectively integrating different pulse sequences is important to optimize performance. However, the best way to integrate different pulse sequences remains unclear. In addition, networks trained with a certain subset of pulse sequences as input are unable to perform when given a subset of those pulse sequences. In this study, we evaluate multiple architectural features and characterize their effects in the task of metastasis segmentation while creating a method to robustly train a network to be able to work given any strict subset of the pulse sequences available during training. Methods: We use a 2.5D DeepLabv3 segmentation network to segment metastases lesions on brain MR's with four pulse sequence inputs. To study how we can best integrate MR pulse sequences for this task, we consider (1) different pulse sequence integration schemas, combining our features at early, middle, and late points within a deep network, (2) different modes of weight sharing for parallel network branches, and (3) a novel integration level dropout layer, which will allow the networks to be robust to performing inference on input with only a subset of pulse sequences available at the training. Results: We find that levels of integration and modes of weight sharing that favor low variance work best in our regime of small amounts of training data (n = 100). By adding an input‐level dropout layer, we could preserve the overall performance of these networks while allowing for inference on inputs with missing pulse sequences. We illustrate not only the generalizability of the network but also the utility of this robustness when applying the trained model to data from a different center, which does not use the same pulse sequences. Finally, we apply network visualization methods to better understand which input features are most important for network performance. Conclusions: Together, these results provide a framework for building networks with enhanced robustness to missing data while maintaining comparable performance in medical imaging applications. [ABSTRACT FROM AUTHOR]

Abstract Background The heart’s energy demand per gram of tissue is the body’s highest and creatine kinase (CK) metabolism, its primary energy reserve, is compromised in common heart diseases. Here, neural-network analysis is used to test whether noninvasive phosphorus (31P) cardiovascular magnetic resonance spectroscopy (CMRS) measurements of cardiac adenosine triphosphate (ATP) energy, phosphocreatine (PCr), the first-order CK reaction rate kf, and the rate of ATP synthesis through CK (CK flux), can predict specific human heart disease and clinical severity. Methods The data comprised the extant 178 complete sets of PCr and ATP concentrations, kf, and CK flux data from human CMRS studies performed on clinical 1.5 and 3 Tesla scanners. Healthy subjects and patients with nonischemic cardiomyopathy, dilated (DCM) or hypertrophic disease, New York Heart Association (NYHA) class I-IV heart failure (HF), or with anterior myocardial infarction are included. Three-layer neural-networks were created to classify disease and to differentiate DCM, hypertrophy and clinical NYHA class in HF patients using leave-one-out training. Network performance was assessed using ‘confusion matrices’ and ‘area-under-the-curve’ (AUC) analyses of ‘receiver operating curves’. Possible methodological bias and network imbalance were tested by segregating 1.5 and 3 Tesla data, and by data augmentation by random interpolation of nearest neighbors, respectively. Results The network differentiated healthy, HF and non-HF cardiac disease with an overall accuracy of 84% and AUC > 90% for each category using the four CK metabolic parameters, alone. HF patients with DCM, hypertrophy, and different NYHA severity were differentiated with ~ 80% overall accuracy independent of CMRS methodology. Conclusions While sample-size was limited in some sub-classes, a neural network classifier applied to noninvasive cardiac 31P CMRS data, could serve as a metabolic biomarker for common disease types and HF severity with clinically-relevant accuracy. Moreover, the network’s ability to individually classify disease and HF severity using CK metabolism alone, implies an intimate relationship between CK metabolism and disease, with subtle underlying phenotypic differences that enable their differentiation. Trial registration ClinicalTrials.gov Identifier: NCT00181259.

Assessment in simulating river flows for a river system can be implemented simultaneously. In general, the majority of the researchers emphasize forecasting on single output for a river system. The present study investigates the applicability and capability of coactive neuro-fuzzy inference system (CANFIS) for simultaneous river flow forecasting for a Barak river system in Assam, India. Besides, two other hybrid model approaches were developed by optimizing the parameters of CANFIS to adopt model's consistency toward achieving more precise and sensitive result which includes the combination of the CANFIS using genetic algorithm (CANFIS-GA) and CANFIS using firefly algorithm (CANFIS-FA). In total, 19,728 sets of recorded hourly concurrent flows data have been collected from different gauging sites pertaining to monsoon seasons. The results of the models (CANFIS, CANFIS-GA and CANFIS-FA) are evaluated, and the best-fit forecasting model(s) is determined using various statistical performance criterions. Also, this study witnessed the significant improvement in the quality of flow forecasting of traditional CANFIS when integrated using metaheuristics algorithms GA and FA. Besides, performance comparisons of the models are made using the artificial neural networks and probabilistics neural networks. In overall, results suggested that CANFIS-FA considerably improved upon other models and provides more better and accurate results for simultaneous flow forecasting in a river system. [ABSTRACT FROM AUTHOR]

The identification of small scale faults (SSFs) and fractures provides an improved understanding of geologic structural features and can be exploited for future drilling prospects. Conventional SSF and fracture characterization are challenging and time-consuming. Thus, the current study was conducted with the following aims: (a) to provide an effective way of utilizing the seismic data in the absence of image logs and cores for characterizing SSFs and fractures; (b) to present an unconventional way of data conditioning using geostatistical and structural filtering; (c) to provide an advanced workflow through multi-attributes, neural networks, and ant-colony optimization (ACO) for the recognition of fracture networks; and (d) to identify the fault and fracture orientation parameters within the study area. Initially, a steering cube was generated, and a dip-steered median filter (DSMF), a dip-steered diffusion filter (DSDF), and a fault enhancement filter (FEF) were applied to sharpen the discontinuities. Multiple structural attributes were applied and shortlisted, including dip and curvature attributes, filtered and unfiltered similarity attributes, thinned fault likelihood (TFL), fracture density, and fracture proximity. These shortlisted attributes were computed through unsupervised vector quantization (UVQ) neural networks. The results of the UVQ revealed the orientations, locations, and extensions of fractures in the study area. The ACO proved helpful in identifying the fracture parameters such as fracture length, dip angle, azimuth, and surface area. The adopted workflow also revealed a small scale fault which had an NNW–SSE orientation with minor heave and throw. The implemented workflow of structural interpretation is helpful for the field development of the study area and can be applied worldwide in carbonate, sand, coal, and shale gas fields. [ABSTRACT FROM AUTHOR]

Background: The heart's energy demand per gram of tissue is the body's highest and creatine kinase (CK) metabolism, its primary energy reserve, is compromised in common heart diseases. Here, neural-network analysis is used to test whether noninvasive phosphorus (31P) cardiovascular magnetic resonance spectroscopy (CMRS) measurements of cardiac adenosine triphosphate (ATP) energy, phosphocreatine (PCr), the first-order CK reaction rate kf, and the rate of ATP synthesis through CK (CK flux), can predict specific human heart disease and clinical severity. Methods: The data comprised the extant 178 complete sets of PCr and ATP concentrations, kf, and CK flux data from human CMRS studies performed on clinical 1.5 and 3 Tesla scanners. Healthy subjects and patients with nonischemic cardiomyopathy, dilated (DCM) or hypertrophic disease, New York Heart Association (NYHA) class I-IV heart failure (HF), or with anterior myocardial infarction are included. Three-layer neural-networks were created to classify disease and to differentiate DCM, hypertrophy and clinical NYHA class in HF patients using leave-one-out training. Network performance was assessed using 'confusion matrices' and 'area-under-the-curve' (AUC) analyses of 'receiver operating curves'. Possible methodological bias and network imbalance were tested by segregating 1.5 and 3 Tesla data, and by data augmentation by random interpolation of nearest neighbors, respectively. Results: The network differentiated healthy, HF and non-HF cardiac disease with an overall accuracy of 84% and AUC > 90% for each category using the four CK metabolic parameters, alone. HF patients with DCM, hypertrophy, and different NYHA severity were differentiated with ~ 80% overall accuracy independent of CMRS methodology. Conclusions: While sample-size was limited in some sub-classes, a neural network classifier applied to noninvasive cardiac 31P CMRS data, could serve as a metabolic biomarker for common disease types and HF severity with clinically-relevant accuracy. Moreover, the network's ability to individually classify disease and HF severity using CK metabolism alone, implies an intimate relationship between CK metabolism and disease, with subtle underlying phenotypic differences that enable their differentiation. Trial registration: ClinicalTrials.gov Identifier: NCT00181259. [ABSTRACT FROM AUTHOR]

Objective: Segmentation of anatomical structures in ultrasound images requires vast radiological knowledge and experience. Moreover, the manual segmentation often results in subjective variations, therefore, an automatic segmentation is desirable. We aim to develop a fully convolutional neural network (FCNN) with attentional deep supervision for the automatic and accurate segmentation of the ultrasound images. Method: FCNN/CNNs are used to infer high-level context using low-level image features. In this paper, a sub-problem specific deep supervision of the FCNN is performed. The attention of fine resolution layers is steered to learn object boundary definitions using auxiliary losses, whereas coarse resolution layers are trained to discriminate object regions from the background. Furthermore, a customized scheme for downweighting the auxiliary losses and a trainable fusion layer are introduced. This produces an accurate segmentation and helps in dealing with the broken boundaries, usually found in the ultrasound images. Results: The proposed network is first tested for blood vessel segmentation in liver images. It results in $F1$ score, mean intersection over union, and dice index of 0.83, 0.83, and 0.79, respectively. The best values observed among the existing approaches are produced by U-net as 0.74, 0.81, and 0.75, respectively. The proposed network also results in dice index value of 0.91 in the lumen segmentation experiments on MICCAI 2011 IVUS challenge dataset, which is near to the provided reference value of 0.93. Furthermore, the improvements similar to vessel segmentation experiments are also observed in the experiment performed to segment lesions. Conclusion: Deep supervision of the network based on the input-output characteristics of the layers results in improvement in overall segmentation accuracy. Significance: Sub-problem specific deep supervision for ultrasound image segmentation is the main contribution of this paper. Currently the network is trained and tested for fixed size inputs. It requires image resizing and limits the performance in small size images. [ABSTRACT FROM AUTHOR]

Humans can learn to perform multiple tasks in succession over the lifespan ("continual" learning), whereas current machine learning systems fail. Here, we investigated the cognitive mechanisms that permit successful continual learning in humans and harnessed our behavioral findings for neural network design. Humans categorized naturalistic images of trees according to one of two orthogonal task rules that were learned by trial and error. Training regimes that focused on individual rules for prolonged periods (blocked training) improved human performance on a later test involving randomly interleaved rules, compared with control regimes that trained in an interleaved fashion. Analysis of human error patterns suggested that blocked training encouraged humans to form "factorized" representation that optimally segregated the tasks, especially for those individuals with a strong prior bias to represent the stimulus space in a well-structured way. By contrast, standard supervised deep neural networks trained on the same tasks suffered catastrophic forgetting under blocked training, due to representational interference in the deeper layers. However, augmenting deep networks with an unsupervised generative model that allowed it to first learn a good embedding of the stimulus space (similar to that observed in humans) reduced catastrophic forgetting under blocked training. Building artificial agents that first learn a model of the world may be one promising route to solving continual task performance in artificial intelligence research. [ABSTRACT FROM AUTHOR]

Trained recurrent networks are powerful tools for modeling dynamic neural computations. We present a target-based method for modifying the full connectivity matrix of a recurrent network to train it to perform tasks involving temporally complex input/output transformations. The method introduces a second network during training to provide suitable “target” dynamics useful for performing the task. Because it exploits the full recurrent connectivity, the method produces networks that perform tasks with fewer neurons and greater noise robustness than traditional least-squares (FORCE) approaches. In addition, we show how introducing additional input signals into the target-generating network, which act as task hints, greatly extends the range of tasks that can be learned and provides control over the complexity and nature of the dynamics of the trained, task-performing network. [ABSTRACT FROM AUTHOR]

How do body-wave traveltimes constrain the Earth's radial (1-D) seismic structure? Existing 1-D seismological models underpin 3-D seismic tomography and earthquake location algorithms. It is therefore crucial to assess the quality of such 1-D models, yet quantifying uncertainties in seismological models is challenging and thus often ignored. Ideally, quality assessment should be an integral part of the inverse method. Our aim in this study is twofold: (i) we show how to solve a general Bayesian non-linear inverse problem and quantify model uncertainties, and (ii) we investigate the constraint on spherically symmetric P-wave velocity (VP) structure provided by body-wave traveltimes from the EHB bulletin (phases Pn, P, PP and PKP). Our approach is based on artificial neural networks, which are very common in pattern recognition problems and can be used to approximate an arbitrary function. We use a Mixture Density Network to obtain 1-D marginal posterior probability density functions (pdfs), which provide a quantitative description of our knowledge on the individual Earth parameters. No linearization or model damping is required, which allows us to infer a model which is constrained purely by the data.We present 1-D marginal posterior pdfs for the 22 VP parameters and seven discontinuity depths in our model. P-wave velocities in the inner core, outer core and lower mantle are resolved well, with standard deviations of ∼0.2 to 1 per cent with respect to the mean of the posterior pdfs. The maximum likelihoods of VP are in general similar to the corresponding ak135 values, which lie within one or two standard deviations from the posterior means, thus providing an independent validation of ak135 in this part of the radial model. Conversely, the data contain little or no information on P-wave velocity in the D′′ layer, the upper mantle and the homogeneous crustal layers. Further, the data do not constrain the depth of the discontinuities in our model. Using additional phases available in the ISC bulletin, such as PcP, PKKP and the converted phases SP and ScP, may enhance the resolvability of these parameters. Finally, we show how the method can be extended to obtain a posterior pdf for a multidimensional model space. This enables us to investigate correlations between model parameters. [ABSTRACT FROM AUTHOR]

Surface quality is one of the most important indicators of the quality of machined parts. The analytical method of defining the arithmetic mean roughness is not applied in practice due to its complexity and empirical models are applied only for certain values of machining parameters. This paper presents the design and development of artificial neural networks (ANNs) for the prediction of the arithmetic mean roughness, which is one of the most common surface roughness parameters. The dataset used for ANN development were obtained experimentally by machining AA7075 aluminum alloy under various machining conditions. With four factors, each having three levels, the full factorial design considers a total of 81 experiments that have to be carried out. Using input factor-level settings and adopting the Taguchi method, the experiments were reduced from 81 runs to 27 runs through an orthogonal design. In this study we aimed to check how reliable the results of artificial neural networks were when obtained based on a small input-output dataset, as in the case of applying the Taguchi methodology of planning a four-factor and three-level experiment, in which 27 trials were conducted. Furthermore, this paper considers the optimization of machining parameters for minimizing surface roughness in machining AA7075 aluminum alloy. The results show that ANNs can be successfully trained with small data and used to predict the arithmetic mean roughness. The best results were achieved by backpropagation multilayer feedforward neural networks using the BR algorithm for training. [ABSTRACT FROM AUTHOR]

Land-use/land cover (LULC) changes have an impact on land surface temperature (LST) at the local, regional, and global scales. To simulate the LULC and LST changes of the environmentally important area of northern Pakistan, this research focused on spatio-temporal LULC and associated LST changes since 1987 and made predictions to 2047. We classified LULC from Landsat TM and ETM data, using the maximum probability supervised categorization approach. LST was retrieved using the Radiative Transfer Equation (RTE) methodology. Furthermore, we simulated LULC using the integrated approaches of Cellular Automata (CA) and Weighted Evidence (WE) and used a regression model to predict LST. The built-up areas and vegetation have increased by 2.1% and 11% due to a decline in the barren land by −8.5% during the last 30 years. The LULC is expected to increase, particularly the built-up and vegetation classes by 2.74% and 13.66%, respectively, and the barren land would decline by −4.2% by 2047. Consequently, the higher LST classes (i.e., 27 °C to <30 °C and ≥30 °C) soared up by about 25.18% and 34.26%, respectively, during the study period, which would further expand to 30.19% and 14.97% by 2047. The lower LST class (i.e., 12 °C to <21 °C) indicated a downtrend of about −41.29% and would further decrease to −3.13% in the next 30 years. The study findings are useful for planning and management, especially for climatologists, land-use planners, and researchers in sustainable land use with rapid urbanization. [ABSTRACT FROM AUTHOR]