Unsupervised novelty detection for time series using a deep learning approach

In the Smart Homes and IoT devices era, abundant available data offers immense potential for enhancing system intelligence. However, the need for effective anomaly detection models to identify and rectify unusual data and behaviors within Smart Home Systems (SHS) remains a critical challenge. This research delves into the relatively unexplored domain of novelty anomaly detection, particularly in the context of unlabeled datasets. Introducing the novel DeepMaly method, this approach provides a practical tool for SHS developers. Functioning seamlessly in an unsupervised manner, DeepMaly distinguishes between seasonal and actual anomalies through a unique process of training on unlabeled pristine features extracted from time series data. Leveraging a combination of Long Short-Term Memory (LSTM) and Deep Convolutional Neural Network (DCNN), the model is primed to detect anomalies in real-time. The research culminates in a comprehensive data prediction and classification process into normal and abnormal data based on specified anomaly thresholds and fraction percentages. Notably, this function operates seamlessly unsupervised, eliminating the need for labeled datasets. The study concludes with a complete data forecasting and sorting method that divides data into normal and abnormal categories based on defined anomaly thresholds and fraction percentages. Working in an unsupervised mode reduces the requirement for labeled datasets. The results highlight the model's prowess in new detection, which has been successfully applied to benchmark datasets. However, there is a restriction since deep learning algorithms can recognize noise as abnormalities. Finally, the investigation enhances SHS anomaly detection, providing a crucial tool for real-time anomaly identification in the ever-changing IoT and Smart Homes scene.

Enhancing subsurface contamination assessment via ensemble prediction of ground electrical property: A Colorado AMD-impacted wetland case study

Acid mine drainage (AMD) is recognized as a major environmental challenge in the Western United States, particularly in Colorado, leading to extreme subsurface contamination issue. Given Colorado's arid climate and dependence on groundwater, an accurate assessment of AMD-induced contamination is deemed crucial. While in past, machine learning (ML)-based inversion algorithms were used to reconstruct ground electrical properties (GEP) such as relative dielectric permittivity (RDP) from ground penetrating radar (GPR) data for contamination assessment, their inherent non-linear nature can introduce significant uncertainty and non-uniqueness into the reconstructed models. This is a challenge that traditional ML methods are not explicitly designed to address. In this study, a probabilistic hybrid technique has been introduced that combines the DeepLabv3+ architecture-based deep convolutional neural network (DCNN) with an ensemble prediction-based Monte Carlo (MC) dropout method. Different MC dropout rates (1%, 5%, and 10%) were initially evaluated using 1D and 2D synthetic GPR data for accurate and reliable RDP model prediction. The optimal rate was chosen based on minimal prediction uncertainty and the closest alignment of the mean or median model with the true RDP model. Notably, with the optimal MC dropout rate, prediction accuracy of over 95% for the 1D and 2D cases was achieved. Motivated by these results, the hybrid technique was applied to field GPR data collected over an AMD-impacted wetland near Silverton, Colorado. The field results underscored the hybrid technique's ability to predict an accurate subsurface RDP distribution for estimating the spatial extent of AMD-induced contamination. Notably, this technique not only provides a precise assessment of subsurface contamination but also ensures consistent interpretations of subsurface condition by different environmentalists examining the same GPR data. In conclusion, the hybrid technique presents a promising avenue for future environmental studies in regions affected by AMD or other contaminants that alter the natural distribution of GEP.

Prediction of brain sex from EEG: using large-scale heterogeneous dataset for developing a highly accurate and interpretable ML model

This study presents a comprehensive examination of sex-related differences in resting-state electroencephalogram (EEG) data, leveraging two different types of machine learning models to predict an individual's sex. We utilized data from the Two Decades-Brainclinics Research Archive for Insights in Neurophysiology (TDBRAIN) EEG study, affirming that gender prediction can be attained with noteworthy accuracy. The best performing model achieved an accuracy of 85% and an ROC AUC of 89%, surpassing all prior benchmarks set using EEG data and rivaling the top-tier results derived from fMRI studies. A comparative analysis of LightGBM and Deep Convolutional Neural Network (DCNN) models revealed DCNN's superior performance, attributed to its ability to learn complex spatial-temporal patterns in the EEG data and handle large volumes of data effectively. Despite this, interpretability remained a challenge for the DCNN model. The LightGBM interpretability analysis revealed that the most important EEG features for accurate sex prediction were related to left fronto-central and parietal EEG connectivity. We also showed the role of both low (delta and theta) and high (beta and gamma) activity in the accurate sex prediction. These results, however, have to be approached with caution, because it was obtained from a dataset comprised largely of participants with various mental health conditions, which limits the generalizability of the results and necessitates further validation in future studies. . Overall, the study illuminates the potential of interpretable machine learning for sex prediction, alongside highlighting the importance of considering individual differences in prediction sex from brain activity.

Multi-Domain Rapid Enhancement Networks for Underwater Images

Images captured during marine engineering operations suffer from color distortion and low contrast. Underwater image enhancement helps to alleviate these problems. Many deep learning models can infer multi-source data, where images with different perspectives exist from multiple sources. To this end, we propose a multichannel deep convolutional neural network (MDCNN) linked to a VGG that can target multi-source (multi-domain) underwater image enhancement. The designed MDCNN feeds data from different domains into separate channels and implements parameters by linking VGGs, which improves the domain adaptation of the model. In addition, to optimize performance, multi-domain image perception loss functions, multilabel soft edge loss for specific image enhancement tasks, pixel-level loss, and external monitoring loss for edge sharpness preprocessing are proposed. These loss functions are set to effectively enhance the structural and textural similarity of underwater images. A series of qualitative and quantitative experiments demonstrate that our model is superior to the state-of-the-art Shallow UWnet in terms of UIQM, and the performance evaluation conducted on different datasets increased by 0.11 on average.

Distinct patterns of neural response to faces from different races in humans and deep networks

Social categories such as the race or ethnicity of an individual are typically conveyed by the visual appearance of the face. The aim of this study was to explore how these differences in facial appearance are represented in human and artificial neural networks. First, we compared the similarity of faces from different races using a neural network trained to discriminate identity. We found that the differences between races were most evident in the fully connected layers of the network. Although these layers were also able to predict behavioural judgements of face identity from human participants, performance was biased toward White faces. Next, we measured the neural response in face-selective regions of the human brain to faces from different races in Asian and White participants. We found distinct patterns of response to faces from different races in face-selective regions. We also found that the spatial pattern of response was more consistent across participants for own-race compared to other-race faces. Together, these findings show that faces from different races elicit different patterns of response in human and artificial neural networks. These differences may underlie the ability to make categorical judgements and explain the behavioural advantage for the recognition of own-race faces.

Exploring the role of texture features in deep convolutional neural networks: Insights from Portilla-Simoncelli statistics

It is well-understood that the performance of Deep Convolutional Neural Networks (DCNNs) in image recognition tasks is influenced not only by shape but also by texture information. Despite this, understanding the internal representations of DCNNs remains a challenging task. This study employs a simplified version of the Portilla-Simoncelli Statistics, termed "minPS," to explore how texture information is represented in a pre-trained VGG network. Using minPS features extracted from texture images, we perform a sparse regression on the activations across various channels in VGG layers. Our findings reveal that channels in the early to middle layers of the VGG network can be effectively described by minPS features. Additionally, we observe that the explanatory power of minPS sub-groups evolves as one ascends the network hierarchy. Specifically, sub-groups termed Linear Cross Scale (LCS) and Energy Cross Scale (ECS) exhibit weak explanatory power for VGG channels. To investigate the relationship further, we compare the original texture images with their synthesized counterparts, generated using VGG, in terms of minPS features. Our results indicate that the absence of certain minPS features suggests their non-utilization in VGG's internal representations.

EMDL_m6Am: identifying N6,2'-O-dimethyladenosine sites based on stacking ensemble deep learning

BACKGROUND

N6, 2'-O-dimethyladenosine (m6Am) is an abundant RNA methylation modification on vertebrate mRNAs and is present in the transcription initiation region of mRNAs. It has recently been experimentally shown to be associated with several human disorders, including obesity genes, and stomach cancer, among others. As a result, N6,2'-O-dimethyladenosine (m6Am) site will play a crucial part in the regulation of RNA if it can be correctly identified.

RESULTS

This study proposes a novel deep learning-based m6Am prediction model, EMDL_m6Am, which employs one-hot encoding to expressthe feature map of the RNA sequence and recognizes m6Am sites by integrating different CNN models via stacking. Including DenseNet, Inflated Convolutional Network (DCNN) and Deep Multiscale Residual Network (MSRN), the sensitivity (Sn), specificity (Sp), accuracy (ACC), Mathews correlation coefficient (MCC) and area under the curve (AUC) of our model on the training data set reach 86.62%, 88.94%, 87.78%, 0.7590 and 0.8778, respectively, and the prediction results on the independent test set are as high as 82.25%, 79.72%, 80.98%, 0.6199, and 0.8211.

CONCLUSIONS

In conclusion, the experimental results demonstrated that EMDL_m6Am greatly improved the predictive performance of the m6Am sites and could provide a valuable reference for the next part of the study. The source code and experimental data are available at: https://github.com/13133989982/EMDL-m6Am .

HRFSVM: identification of fish disease using hybrid Random Forest and Support Vector Machine

Aquaculture fish diseases pose a serious threat to the security of food supplies. Fish species vary widely, and because they resemble one another so much, it is challenging to distinguish between them based solely on appearance. To stop the spread of disease, it is important to identify sick fish as soon as possible. Due to a lack of necessary infrastructure, it is still difficult to identify infected fish in aquaculture at an early stage. It is essential to promptly identify sick fish to stop the spread of disease. The purpose of this work is to suggest a machine learning technique based on the DCNN method for identifying and categorizing fish diseases. To solve problems involving global optimization, this paper suggests a brand-new hybrid algorithm called the Whale Optimization Algorithm with Genetic Algorithm (WOA-GA) and Ant Colony Optimization. In this work, for classification, the hybrid Random Forest algorithm is used. To facilitate the increased quality, distinctions between both the proposed WOA-GA-based DCNN architecture and the presently used methods for machine learning have been made. The effectiveness of the proposed detection technique is done with MATLAB. Performance metrics like sensitivity, specificity, accuracy, precision, recall, F-measure, NPV, FPR, FNR, and MCC are compared to the performance of the proposed technique.

Flamingo-Optimization-Based Deep Convolutional Neural Network for IoT-Based Arrhythmia Classification

Cardiac arrhythmia is a deadly disease that threatens the lives of millions of people, which shows the need for earlier detection and classification. An abnormal signal in the heart causing arrhythmia can be detected at an earlier stage when the health data from the patient are monitored using IoT technology. Arrhythmias may suddenly lead to death and the classification of arrhythmias is considered a complicated process. In this research, an effective classification model for the classification of heart disease is developed using flamingo optimization. Initially, the ECG signal from the heart is collected and then it is subjected to the preprocessing stage; to detect and control the electrical activity of the heart, the electrocardiogram (ECG) is used. The input signals collected using IoT nodes are collectively presented in the base station for the classification using flamingo-optimization-based deep convolutional networks, which effectively predict the disease. With the aid of communication technologies and the contribution of IoT, medical professionals can easily monitor the health condition of patients. The performance is analyzed in terms of accuracy, sensitivity, and specificity.

A deep learning approach for automatic identification of ancient agricultural water harvesting systems

Despite the harsh climatic conditions in the Central Negev Desert, Israel, thousands of dry stonewalls were built across ephemeral streams between the fourth and seventh centuries CE to sustain productive agricultural activity. Since 640 CE, many of these ancient terraces have remained untouched but buried by sediments, covered by natural vegetation, and partially destroyed. The main goal of the current research is to develop a procedure for the automatic recognition of ancient water harvesting systems by incorporating two remote sensing datasets (a high-resolution color orthophoto and LiDAR-derived topographic variables) and two advanced processing methods (an object-based image analysis (OBIA) and a deep convolutional neural networks (DCNN) model). A confusion matrix of object-based classification revealed an overall accuracy of 86% and a Kappa coefficient of 0.79. The DCNN model achieved a Mean Intersection over Union (MIoU) value for testing datasets of 53. The individual IoU values of terraces and sidewalls were 33.2 and 30.1, respectively. The current study demonstrates how incorporating OBIA, aerial photographs, and LiDAR in the context of DCNN improves the identification and mapping of archaeological structures.

A CAD System for Lung Cancer Detection Using Hybrid Deep Learning Techniques

Lung cancer starts and spreads in the tissues of the lungs, more specifically, in the tissue that forms air passages. This cancer is reported as the leading cause of cancer deaths worldwide. In addition to being the most fatal, it is the most common type of cancer. Nearly 47,000 patients are diagnosed with it annually worldwide. This article proposes a fully automated and practical system to identify and classify lung cancer. This system aims to detect cancer in its early stage to save lives if possible or reduce the death rates. It involves a deep convolutional neural network (DCNN) technique, VGG-19, and another deep learning technique, long short-term memory networks (LSTMs). Both tools detect and classify lung cancers after being customized and integrated. Furthermore, image segmentation techniques are applied. This system is a type of computer-aided diagnosis (CAD). After several experiments on MATLAB were conducted, the results show that this system achieves more than 98.8% accuracy when using both tools together. Various schemes were developed to evaluate the considered disease. Three lung cancer datasets, downloaded from the Kaggle website and the LUNA16 grad challenge, were used to train the algorithm, test it, and prove its correctness. Lastly, a comparative evaluation between the proposed approach and some works from the literature is presented. This evaluation focuses on the four performance metrics: accuracy, recall, precision, and F-score. This system achieved an average of 99.42% accuracy and 99.76, 99.88, and 99.82% for recall, precision, and F-score, respectively, when VGG-19 was combined with LSTMs. In addition, the results of the comparison evaluation show that the proposed algorithm outperforms other methods and produces exquisite findings. This study concludes that this model can be deployed to aid and support physicians in diagnosing lung cancer correctly and accurately. This research reveals that the presented method has functionality, competence, and value among other implemented models.

Discriminating Healthy Optic Discs and Visible Optic Disc Drusen on Fundus Autofluorescence and Color Fundus Photography Using Deep Learning-A Pilot Study

The aim of this study was to use deep learning based on a deep convolutional neural network (DCNN) for automated image classification of healthy optic discs (OD) and visible optic disc drusen (ODD) on fundus autofluorescence (FAF) and color fundus photography (CFP). In this study, a total of 400 FAF and CFP images of patients with ODD and healthy controls were used. A pre-trained multi-layer Deep Convolutional Neural Network (DCNN) was trained and validated independently on FAF and CFP images. Training and validation accuracy and cross-entropy were recorded. Both generated DCNN classifiers were tested with 40 FAF and CFP images (20 ODD and 20 controls). After the repetition of 1000 training cycles, the training accuracy was 100%, the validation accuracy was 92% (CFP) and 96% (FAF), respectively. The cross-entropy was 0.04 (CFP) and 0.15 (FAF). The sensitivity, specificity, and accuracy of the DCNN for classification of FAF images was 100%. For the DCNN used to identify ODD on color fundus photographs, sensitivity was 85%, specificity 100%, and accuracy 92.5%. Differentiation between healthy controls and ODD on CFP and FAF images was possible with high specificity and sensitivity using a deep learning approach.

Deep-Learning-Based Context-Aware Multi-Level Information Fusion Systems for Indoor Mobile Robots Safe Navigation

Hazardous object detection (escalators, stairs, glass doors, etc.) and avoidance are critical functional safety modules for autonomous mobile cleaning robots. Conventional object detectors have less accuracy for detecting low-feature hazardous objects and have miss detection, and the false classification ratio is high when the object is under occlusion. Miss detection or false classification of hazardous objects poses an operational safety issue for mobile robots. This work presents a deep-learning-based context-aware multi-level information fusion framework for autonomous mobile cleaning robots to detect and avoid hazardous objects with a higher confidence level, even if the object is under occlusion. First, the image-level-contextual-encoding module was proposed and incorporated with the Faster RCNN ResNet 50 object detector model to improve the low-featured and occluded hazardous object detection in an indoor environment. Further, a safe-distance-estimation function was proposed to avoid hazardous objects. It computes the distance of the hazardous object from the robot's position and steers the robot into a safer zone using detection results and object depth data. The proposed framework was trained with a custom image dataset using fine-tuning techniques and tested in real-time with an in-house-developed mobile cleaning robot, BELUGA. The experimental results show that the proposed algorithm detected the low-featured and occluded hazardous object with a higher confidence level than the conventional object detector and scored an average detection accuracy of 88.71%.

Deep convolutional neural network with sine cosine algorithm based wastewater treatment systems

Wastewater treatment systems are essential in today's business to meet the ever-increasing requirements of environmental regulations while also limiting the environmental impact of the sector's discharges. A new control and management information system is needed to handle the residual fluids. This study advises that Wastewater Treatment System (WWTS) operators use intelligent technologies that analyze data and forecast the future behaviour of processes. This method incorporates industrial data into the wastewater treatment model. Deep Convolutional Neural Network (DCNN) and Since Cosine Algorithm (SCA), two powerful artificial neural networks, were used to predict these properties over time. Remediation actions can be taken to ensure procedures are carried out in accordance with the specifications. Water treatment facilities can benefit from this technology because of its sophisticated process that changes feature dynamically and inconsistently. The ultimate goal is to improve the precision with which wastewater treatment models create their predictions. Using DCNN and SCA techniques, the Chemical Oxygen Demand (COD) in wastewater treatment system input and effluent is estimated in this study. Finally, the DCNN-SCA model is applied for the optimization, and it assists in improving the predictive performance. The experimental validation of the DCNN-SCA model is tested and the outcomes are investigated under various prospects. The DCNN-SCA model has achieved a maximum accuracy performance and proving that it outperforms compare with the prevailing techniques over recent approaches. The DCNN-SCA-WWTS model has shown maximum performance Under 600 data, DCNN-SCA-WWTS has a precision of 97.63%, a recall of 96.37%, a F score of 95.31%, an accuracy of 96.27%, an RMSE of 27.55%, and a MAPE of 20.97%.

A Model to Predict Heartbeat Rate Using Deep Learning Algorithms

ECG provides critical information in a waveform about the heart's condition. This information is crucial to physicians as it is the first thing to be performed by cardiologists. When COVID-19 spread globally and became a pandemic, the government of Saudi Arabia placed various restrictions and guidelines to protect and save citizens and residents. One of these restrictions was preventing individuals from touching any surface in public and private places. In addition, the authorities placed a mandatory rule in all public facilities and the private sector to evaluate the temperature of individuals before entering. Thus, the idea of this study stems from the need to have a touchless technique to determine heartbeat rate. This article proposes a viable and dependable method to estimate an average heartbeat rate based on the reflected light on the skin. This model uses various deep learning tools, including AlexNet, Convolutional Neural Networks (CNNs), Long Short-Term Memory Networks (LSTMs), and ResNet50V2. Three scenarios have been conducted to evaluate and validate the presented model. In addition, the proposed approach takes its inputs from video streams and converts these streams into frames and images. Numerous trials have been conducted on volunteers to validate the method and assess its outputs in terms of accuracy, mean absolute error (MAE), and mean squared error (MSE). The proposed model achieves an average 99.78% accuracy, MAE is 0.142 when combing LSTMs and ResNet50V2, while MSE is 1.82. Moreover, a comparative measurement between the presented algorithm and some studies from the literature based on utilized methods, MAE, and MSE are performed. The achieved outcomes reveal that the developed technique surpasses other methods. Moreover, the findings show that this algorithm can be applied in healthcare facilities and aid physicians.

Cloud-based COVID-19 disease prediction system from X-Ray images using convolutional neural network on smartphone

COVID-19 has engulfed over 200 nations through human-to-human transmission, either directly or indirectly. Reverse Transcription-polymerase Chain Reaction (RT-PCR) has been endorsed as a standard COVID-19 diagnostic procedure but has caveats such as low sensitivity, the need for a skilled workforce, and is time-consuming. Coronaviruses show significant manifestation in Chest X-Ray (CX-Ray) images and, thus, can be a viable option for an alternate COVID-19 diagnostic strategy. An automatic COVID-19 detection system can be developed to detect the disease, thus reducing strain on the healthcare system. This paper discusses a real-time Convolutional Neural Network (CNN) based system for COVID-19 illness prediction from CX-Ray images on the cloud. The implemented CNN model displays exemplary results, with training accuracy being 99.94% and validation accuracy reaching 98.81%. The confusion matrix was utilized to assess the models' outcome and achieved 99% precision, 98% recall, 99% F1 score, 100% training area under the curve (AUC) and 98.3% validation AUC. The same CX-Ray dataset was also employed to predict the COVID-19 disease with deep Convolution Neural Networks (DCNN), such as ResNet50, VGG19, InceptonV3, and Xception. The prediction outcome demonstrated that the present CNN was more capable than the DCNN models. The efficient CNN model was deployed to the Platform as a Service (PaaS) cloud.

DeSa COVID-19: Deep salient COVID-19 image-based quality assessment

This study offers an advanced method to evaluate the coronavirus disease 2019 (COVID-19) image quality. The salient COVID-19 image map is incorporated with the deep convolutional neural network (DCNN), namely DeSa COVID-19, which exerts the n-convex method for the full-reference image quality assessment (FR-IQA). The glaring outcomes substantiate that DeSa COVID-19 and the recommended DCNN architecture can convey a remarkable accomplishment on the COVID-chestxray and the COVID-CT datasets, respectively. The salient COVID-19 image map is also gauged in the minuscule COVID-19 image patches. The exploratory results attest that DeSa COVID-19 and the recommended DCNN methods are very good accomplishment compared with other advanced methods on COVID-chestxray and COVID-CT datasets, respectively. The recommended DCNN also acquires the enhanced outgrowths faced with several advanced full-reference-medical-image-quality-assessment (FR-MIQA) techniques in the fast fading (FF), blocking artifact (BA), white noise Gaussian (WG), JPEG, and JPEG2000 (JP2K) in the distorted and undistorted COVID-19 images. The Spearman's rank order correlation coefficient (SROCC) and the linear correlation coefficient (LCC) appraise the recommended DCNN and DeSa COVID-19 fulfillment which are compared the recent FR-MIQA methods. The DeSa COVID-19 evaluation outshines 2.63 % and 2.62 % higher compared the recommended DCNN, and 28.53 % and 29.01 % esteem all of advanced FR-MIQAs methods on SROCC and LCC measures, respectively. The shift add operations of trigonometric, logarithmic, and exponential functions are mowed down in the computational complexity of the DeSa COVID-19 and the recommended DCNN. The DeSa COVID-19 more superior the recommended DCNN and also the other recent full-reference medical image quality assessment methods.

Multiparameter optimization system with DCNN in precision agriculture for advanced irrigation planning and scheduling based on soil moisture estimation

Agriculture is a distinct sector of a country's economy. In recent years, new patterns have evolved in the agricultural industry. In conjunction with sensor scaling down and precision agriculture, the field of remote sensor networks, such as the wireless sensor network (WSN), was developed. Its major purpose is to make horticultural operations simpler to identify, assess, and manage. This paper uses the proposed DCNN to predict soil moisture and plan irrigation for precision agriculture farmers to reduce water consumption used for cultivation and increase production yield by comparing water content during various stages of plant growth and integrating IoT applications into agriculture. It also optimizes the water level for future irrigation decisions to maintain crop growth and water stability. The data must be served and stored in the form of a grid view, according to Apriori and GRU (gated recurrent unit). Using numerous sensor and parameter modelling methodologies, this system assists in the prediction of irrigation planning based on irrigation needs. The predicted parameters include soil moisture, temperature, and humidity. This observed experimental data supports smart irrigation in crop production with a high yield and little water use. DCNN has a 98.5% experimental result accuracy rate and the MSE value is predicted in DCNN 99.25% of the time.

An Enhanced Scheme for Reducing the Complexity of Pointwise Convolutions in CNNs for Image Classification Based on Interleaved Grouped Filters without Divisibility Constraints

In image classification with Deep Convolutional Neural Networks (DCNNs), the number of parameters in pointwise convolutions rapidly grows due to the multiplication of the number of filters by the number of input channels that come from the previous layer. Existing studies demonstrated that a subnetwork can replace pointwise convolutional layers with significantly fewer parameters and fewer floating-point computations, while maintaining the learning capacity. In this paper, we propose an improved scheme for reducing the complexity of pointwise convolutions in DCNNs for image classification based on interleaved grouped filters without divisibility constraints. The proposed scheme utilizes grouped pointwise convolutions, in which each group processes a fraction of the input channels. It requires a number of channels per group as a hyperparameter Ch. The subnetwork of the proposed scheme contains two consecutive convolutional layers K and L, connected by an interleaving layer in the middle, and summed at the end. The number of groups of filters and filters per group for layers K and L is determined by exact divisions of the original number of input channels and filters by Ch. If the divisions were not exact, the original layer could not be substituted. In this paper, we refine the previous algorithm so that input channels are replicated and groups can have different numbers of filters to cope with non exact divisibility situations. Thus, the proposed scheme further reduces the number of floating-point computations (11%) and trainable parameters (10%) achieved by the previous method. We tested our optimization on an EfficientNet-B0 as a baseline architecture and made classification tests on the CIFAR-10, Colorectal Cancer Histology, and Malaria datasets. For each dataset, our optimization achieves a saving of 76%, 89%, and 91% of the number of trainable parameters of EfficientNet-B0, while keeping its test classification accuracy.

SAS-SEINet: A SNR-Aware Adaptive Scalable SEI Neural Network Accelerator Using Algorithm-Hardware Co-Design for High-Accuracy and Power-Efficient UAV Surveillance

As a potential air control measure, RF-based surveillance is one of the most commonly used unmanned aerial vehicles (UAV) surveillance methods that exploits specific emitter identification (SEI) technology to identify captured RF signal from ground controllers to UAVs. Recently many SEI algorithms based on deep convolution neural network (DCNN) have emerged. However, there is a lack of the implementation of specific hardware. This paper proposes a high-accuracy and power-efficient hardware accelerator using an algorithm-hardware co-design for UAV surveillance. For the algorithm, we propose a scalable SEI neural network with SNR-aware adaptive precision computation. With SNR awareness and precision reconfiguration, it can adaptively switch between DCNN and binary DCNN to cope with low SNR and high SNR tasks, respectively. In addition, a short-time Fourier transform (STFT) reusing DCNN method is proposed to pre-extract feature of UAV signal. For hardware, we designed a SNR sensing engine, denoising engine, and specialized DCNN engine with hybrid-precision convolution and memory access, aiming at SEI acceleration. Finally, we validate the effectiveness of our design on a FPGA, using a public UAV dataset. Compared with a state-of-the-art algorithm, our method can achieve the highest accuracy of 99.3% and an F1 score of 99.3%. Compared with other hardware designs, our accelerator can achieve the highest power efficiency of 40.12 Gops/W and 96.52 Gops/W with INT16 precision and binary precision.

ECG signal classification based on deep CNN and BiLSTM

Background

Currently, cardiovascular disease has become a major disease endangering human health, and the number of such patients is growing. Electrocardiogram (ECG) is an important basis for {medical doctors to diagnose the cardiovascular disease, which can truly reflect the health of the heart. In this context, the contradiction between the lack of medical resources and the surge in the number of patients has become increasingly prominent. The use of computer-aided diagnosis of cardiovascular disease has become particularly important, so the study of ECG automatic classification method has a strong practical significance.

Methods

This article proposes a new method for automatic identification and classification of ECG.We have developed a dense heart rhythm network that combines a 24-layer Deep Convolutional Neural Network (DCNN) and Bidirectional Long Short-Term Memory (BiLSTM) to deeply mine the hierarchical and time-sensitive features of ECG data. Three different sizes of convolution kernels (32, 64 and 128) are used to mine the detailed features of the ECG signal, and the original ECG is filtered using a combination of wavelet transform and median filtering to eliminate the influence of noise on the signal. A new loss function is proposed to control the fluctuation of loss during the training process, and convergence mapping of the tan function in the range of 0–1 is employed to better reflect the model training loss and correct the optimization direction in time.

Results

We applied the dataset provided by the 2017 PhysioNet/CINC challenge for verification. The experiment adopted ten-fold cross validation,and obtained an accuracy rate of 89.3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\%$$\end{document}% and an F1 score of 0.891.

Conclusions

This article proposes its own method in the aspects of ECG data preprocessing, feature extraction and loss function design. Compared with the existing methods, this method improves the accuracy of automatic ECG classification and is helpful for clinical diagnosis and self-monitoring of atrial fibrillation.

Residual refinement for interactive skin lesion segmentation

BACKGROUND

Image segmentation is a difficult and classic problem. It has a wide range of applications, one of which is skin lesion segmentation. Numerous researchers have made great efforts to tackle the problem, yet there is still no universal method in various application domains.

RESULTS

We propose a novel approach that combines a deep convolutional neural network with a grabcut-like user interaction to tackle the interactive skin lesion segmentation problem. Slightly deviating from grabcut user interaction, our method uses boxes and clicks. In addition, contrary to existing interactive segmentation algorithms that combine the initial segmentation task with the following refinement task, we explicitly separate these tasks by designing individual sub-networks. One network is SBox-Net, and the other is Click-Net. SBox-Net is a full-fledged segmentation network that is built upon a pre-trained, state-of-the-art segmentation model, while Click-Net is a simple yet powerful network that combines feature maps extracted from SBox-Net and user clicks to residually refine the mistakes made by SBox-Net. Extensive experiments on two public datasets, PH2 and ISIC, confirm the effectiveness of our approach.

CONCLUSIONS

We present an interactive two-stage pipeline method for skin lesion segmentation, which was demonstrated to be effective in comprehensive experiments.

Automatic morphological classification of mitral valve diseases in echocardiographic images based on explainable deep learning methods

PURPOSE

Carpentier's functional classification is a guide to explain the types of mitral valve regurgitation based on morphological features. There are four types of pathological morphologies, regardless of the presence or absence of mitral regurgitation: Type I, normal; Type II, mitral valve prolapse; Type IIIa, mitral valve stenosis; and Type IIIb, restricted mitral leaflet motion. The aim of this study was to automatically classify mitral valves using echocardiographic images.

METHODS

In our procedure, after the classification of apical 4-chamber (A4C) and parasternal long-axis (PLA) views, we extracted the systolic/diastolic phase of the cardiac cycle by calculating the left ventricular area. Six typical pre-trained models were fine-tuned with a 4-class model for the PLA and a 3-class model for the A4C views. As an additional contribution, to provide explainability, we applied the Gradient-weighted Class Activation Mapping (Grad-CAM) algorithm to visualize areas of echocardiographic images where the different models generated a prediction.

RESULTS

This approach conferred a proper understanding of where various networks "look" into echocardiographic images to predict the four types of pathological mitral valve morphologies. Considering the accuracy metric and Grad-CAM maps and by applying the Inception-ResNet-v2 architecture to classify Type II in the PLA view and ResNeXt50 architecture to classify the other three classes in the A4C view, we achieved an 80% rate of model accuracy in the test data set.

CONCLUSIONS

We suggest an explainable, fully automated, and rule-based procedure to classify the four types of mitral valve morphologies based on Carpentier's functional classification using deep learning on transthoracic echocardiographic images. Our study results infer the feasibility of the use of deep learning models to prepare quick and precise assessments of mitral valve morphologies in echocardiograms. According to our knowledge, our study is the first one that provides a public data set regarding the Carpentier classification of MV pathologies.

A multi model ensemble based deep convolution neural network structure for detection of COVID19

The year 2020 will certainly be remembered for the COVID-19 outbreak. First reported in Wuhan city of China back in December 2019, the number of people getting affected by this contagious virus has grown exponentially. Given the population density of India, the implementation of the mantra of the test, track, and isolate is not obtaining satisfactory results. A shortage of testing kits and an increasing number of fresh cases encouraged us to come up with a model that can aid radiologists in detecting COVID19 using chest Xray images. In the proposed framework the low level features from the Chest X-ray images are extracted using an ensemble of four pre-trained Deep Convolutional Neural Network (DCNN) architectures, namely VGGNet, GoogleNet, DenseNet, and NASNet and later on are fed to a fully connected layer for classification. The proposed multi model ensemble architecture is validated on two publicly available datasets and one private dataset. We have shown that our multi model ensemble architecture performs better than single classifier. On the publicly available dataset we have obtained an accuracy of 88.98% for three class classification and for binary class classification we report an accuracy of 98.58%. Validating the performance on private dataset we obtained an accuracy of 93.48%. The source code and the dataset are made available in the github linkhttps://github.com/sagardeepdeb/ensemble-model-for-COVID-detection.

Tactile Object Recognition for Humanoid Robots Using New Designed Piezoresistive Tactile Sensor and DCNN

A tactile sensor array is a crucial component for applying physical sensors to a humanoid robot. This work focused on developing a palm-size tactile sensor array (56.0 mm × 56.0 mm) to apply object recognition for the humanoid robot hand. This sensor was based on a PCB technology operating with the piezoresistive principle. A conductive polymer composites sheet was used as a sensing element and the matrix array of this sensor was 16 × 16 pixels. The sensitivity of this sensor was evaluated and the sensor was installed on the robot hand. The tactile images, with resolution enhancement using bicubic interpolation obtained from 20 classes, were used to train and test 19 different DCNNs. InceptionResNetV2 provided superior performance with 91.82% accuracy. However, using the multimodal learning method that included InceptionResNetV2 and XceptionNet, the highest recognition rate of 92.73% was achieved. Moreover, this recognition rate improved when the object exploration was applied to demonstrate.

Road Extraction from High Resolution Remote Sensing Images Based on Vector Field Learning

Accurate and up-to-date road network information is very important for the Geographic Information System (GIS) database, traffic management and planning, automatic vehicle navigation, emergency response and urban pollution sources investigation. In this paper, we use vector field learning to extract roads from high resolution remote sensing imaging. This method is usually used for skeleton extraction in nature image, but seldom used in road extraction. In order to improve the accuracy of road extraction, three vector fields are constructed and combined respectively with the normal road mask learning by a two-task network. The results show that all the vector fields are able to significantly improve the accuracy of road extraction, no matter the field is constructed in the road area or completely outside the road. The highest F1 score is 0.7618, increased by 0.053 compared with using only mask learning.

A deep learning approach for the automatic recognition of prosthetic mitral valve in echocardiographic images

The first step in the automatic evaluation of the cardiac prosthetic valve is the recognition of such valves in echocardiographic images. This research surveyed whether a deep convolutional neural network (DCNN) could improve the recognition of prosthetic mitral valve in conventional 2D echocardiographic images. An efficient intervention to decrease the misreading rate of the prosthetic mitral valve is required for non-expert cardiologists. This intervention could serve as a section of a fully-automated analysis chain, alleviate the cardiologist's workload, and improve precision and time management, especially in an emergent situation. Additionally, it might be suitable for pre-labeling large databases of unclassified images. We, therefore, introduce a large publicly-available annotated dataset for the purpose of prosthetic mitral valve recognition. We utilized 2044 comprehensive non-stress transthoracic echocardiographic studies. Totally, 1597 patients had natural mitral valves and 447 patients had prosthetic valves. Each case contained 1 cycle of echocardiographic images from the apical 4-chamber (A4C) and the parasternal long-axis (PLA) views. Thirteen versions of the state-of-the-art models were independently trained, and the ensemble predictions were performed using those versions. For the recognition of prosthetic mitral valves from natural mitral valves, the area under the receiver-operating characteristic curve (AUC) made by the deep learning algorithm was similar to that made by cardiologists (0.99). In this research, EfficientNetB3 architecture in the A4C view and the EfficientNetB4 architecture in the PLA view were the best models among the other pre-trained DCNN models.

Autosegmentation of Prostate Zones and Cancer Regions from Biparametric Magnetic Resonance Images by Using Deep-Learning-Based Neural Networks

The accuracy in diagnosing prostate cancer (PCa) has increased with the development of multiparametric magnetic resonance imaging (mpMRI). Biparametric magnetic resonance imaging (bpMRI) was found to have a diagnostic accuracy comparable to mpMRI in detecting PCa. However, prostate MRI assessment relies on human experts and specialized training with considerable inter-reader variability. Deep learning may be a more robust approach for prostate MRI assessment. Here we present a method for autosegmenting the prostate zone and cancer region by using SegNet, a deep convolution neural network (DCNN) model. We used PROSTATEx dataset to train the model and combined different sequences into three channels of a single image. For each subject, all slices that contained the transition zone (TZ), peripheral zone (PZ), and PCa region were selected. The datasets were produced using different combinations of images, including T2-weighted (T2W) images, diffusion-weighted images (DWI) and apparent diffusion coefficient (ADC) images. Among these groups, the T2W + DWI + ADC images exhibited the best performance with a dice similarity coefficient of 90.45% for the TZ, 70.04% for the PZ, and 52.73% for the PCa region. Image sequence analysis with a DCNN model has the potential to assist PCa diagnosis.

An integrated deep learning and dynamic programming method for predicting tumor suppressor genes, oncogenes, and fusion from PDB structures

Mutations in proto-oncogenes (ONGO) and the loss of regulatory function of tumor suppression genes (TSG) are the common underlying mechanism for uncontrolled tumor growth. While cancer is a heterogeneous complex of distinct diseases, finding the potentiality of the genes related functionality to ONGO or TSG through computational studies can help develop drugs that target the disease. This paper proposes a classification method that starts with a preprocessing stage to extract the feature map sets from the input 3D protein structural information. The next stage is a deep convolutional neural network stage (DCNN) that outputs the probability of functional classification of genes. We explored and tested two approaches: in Approach 1, all filtered and cleaned 3D-protein-structures (PDB) are pooled together, whereas in Approach 2, the primary structures and their corresponding PDBs are separated according to the genes' primary structural information. Following the DCNN stage, a dynamic programming-based method is used to determine the final prediction of the primary structures' functionality. We validated our proposed method using the COSMIC online database. For the ONGO vs TSG classification problem the AUROC of the DCNN stage for Approach 1 and Approach 2 DCNN are 0.978 and 0.765, respectively. The AUROCs of the final genes' primary structure functionality classification for Approach 1 and Approach 2 are 0.989, and 0.879, respectively. For comparison, the current state-of-the-art reported AUROC is 0.924. Our results warrant further study to apply the deep learning models to humans' (GRCh38) genes, for predicting their corresponding probabilities of functionality in the cancer drivers.

Novel Deep Convolutional Neural Network-Based Contextual Recognition of Arabic Handwritten Scripts

Offline Arabic Handwriting Recognition (OAHR) has recently become instrumental in the areas of pattern recognition and image processing due to its application in several fields, such as office automation and document processing. However, OAHR continues to face several challenges, including high variability of the Arabic script and its intrinsic characteristics such as cursiveness, ligatures, and diacritics, the unlimited variation in human handwriting, and the lack of large public databases. In this paper, we introduce a novel context-aware model based on deep neural networks to address the challenges of recognizing offline handwritten Arabic text, including isolated digits, characters, and words. Specifically, we propose a supervised Convolutional Neural Network (CNN) model that contextually extracts optimal features and employs batch normalization and dropout regularization parameters. This aims to prevent overfitting and further enhance generalization performance when compared to conventional deep learning models. We employ a number of deep stacked-convolutional layers to design the proposed Deep CNN (DCNN) architecture. The model is extensively evaluated and shown to demonstrate excellent classification accuracy when compared to conventional OAHR approaches on a diverse set of six benchmark databases, including MADBase (Digits), CMATERDB (Digits), HACDB (Characters), SUST-ALT (Digits), SUST-ALT (Characters), and SUST-ALT (Names). A further experimental study is conducted on the benchmark Arabic databases by exploiting transfer learning (TL)-based feature extraction which demonstrates the superiority of our proposed model in relation to state-of-the-art VGGNet-19 and MobileNet pre-trained models. Finally, experiments are conducted to assess comparative generalization capabilities of the models using another language database , specifically the benchmark MNIST English isolated Digits database, which further confirm the superiority of our proposed DCNN model.

You won't believe what this guy is doing with the potato: The ObjAct stimulus-set depicting human actions on congruent and incongruent objects

Perception famously involves both bottom-up and top-down processes. The latter are influenced by our previous knowledge and expectations about the world. In recent years, many studies have focused on the role of expectations in perception in general, and in object processing in particular. Yet studying this question is not an easy feat, requiring-among other things-the creation and validation of appropriate stimuli. Here, we introduce the ObjAct stimulus-set of free-to-use, highly controlled real-life scenes, on which critical objects are pasted. All scenes depict human agents performing an action with an object that is either congruent or incongruent with the action. The focus on human actions yields highly constraining contexts, strengthening congruency effects. The stimuli were analyzed for low-level properties, using the SHINE toolbox to control for luminance and contrast, and using a deep convolutional neural network to mimic V1 processing and potentially discover other low-level factors that might differ between congruent and incongruent scenes. Two online validation studies (N = 500) were also conducted to assess the congruency manipulation and collect additional ratings of our images (e.g., arousal, likeability, visual complexity). We also provide full descriptions of the online sources from which all images were taken, as well as verbal descriptions of their content. Taken together, this extensive validation and characterization procedure makes the ObjAct stimulus-set highly informative and easy to use for future researchers in multiple fields, from object and scene processing, through top-down contextual effects, to the study of actions.

Deep learning supported disease detection with multi-modality image fusion

Multi-modal image fusion techniques aid the medical experts in better disease diagnosis by providing adequate complementary information from multi-modal medical images. These techniques enhance the effectiveness of medical disorder analysis and classification of results. This study aims at proposing a novel technique using deep learning for the fusion of multi-modal medical images. The modified 2D Adaptive Bilateral Filters (M-2D-ABF) algorithm is used in the image pre-processing for filtering various types of noises. The contrast and brightness are improved by applying the proposed Energy-based CLAHE algorithm in order to preserve the high energy regions of the multimodal images. Images from two different modalities are first registered using mutual information and then registered images are fused to form a single image. In the proposed fusion scheme, images are fused using Siamese Neural Network and Entropy (SNNE)-based image fusion algorithm. Particularly, the medical images are fused by using Siamese convolutional neural network structure and the entropy of the images. Fusion is done on the basis of score of the SoftMax layer and the entropy of the image. The fused image is segmented using Fast Fuzzy C Means Clustering Algorithm (FFCMC) and Otsu Thresholding. Finally, various features are extracted from the segmented regions. Using the extracted features, classification is done using Logistic Regression classifier. Evaluation is performed using publicly available benchmark dataset. Experimental results using various pairs of multi-modal medical images reveal that the proposed multi-modal image fusion and classification techniques compete the existing state-of-the-art techniques reported in the literature.

End-to-End Automated Latent Fingerprint Identification With Improved DCNN-FFT Enhancement

Automatic Latent Fingerprint Identification Systems (AFIS) are most widely used by forensic experts in law enforcement and criminal investigations. One of the critical steps used in automatic latent fingerprint matching is to automatically extract reliable minutiae from fingerprint images. Hence, minutiae extraction is considered to be a very important step in AFIS. The performance of such systems relies heavily on the quality of the input fingerprint images. Most of the state-of-the-art AFIS failed to produce good matching results due to poor ridge patterns and the presence of background noise. To ensure the robustness of fingerprint matching against low quality latent fingerprint images, it is essential to include a good fingerprint enhancement algorithm before minutiae extraction and matching. In this paper, we have proposed an end-to-end fingerprint matching system to automatically enhance, extract minutiae, and produce matching results. To achieve this, we have proposed a method to automatically enhance the poor-quality fingerprint images using the "Automated Deep Convolutional Neural Network (DCNN)" and "Fast Fourier Transform (FFT)" filters. The Deep Convolutional Neural Network (DCNN) produces a frequency enhanced map from fingerprint domain knowledge. We propose an "FFT Enhancement" algorithm to enhance and extract the ridges from the frequency enhanced map. Minutiae from the enhanced ridges are automatically extracted using a proposed "Automated Latent Minutiae Extractor (ALME)". Based on the extracted minutiae, the fingerprints are automatically aligned, and a matching score is calculated using a proposed "Frequency Enhanced Minutiae Matcher (FEMM)" algorithm. Experiments are conducted on FVC2002, FVC2004, and NIST SD27 latent fingerprint databases. The minutiae extraction results show significant improvement in precision, recall, and F1 scores. We obtained the highest Rank-1 identification rate of 100% for FVC2002/2004 and 84.5% for NIST SD27 fingerprint databases. The matching results reveal that the proposed system outperforms state-of-the-art systems.

The Application of Deep Convolutional Neural Networks to Brain Cancer Images: A Survey

In recent years, improved deep learning techniques have been applied to biomedical image processing for the classification and segmentation of different tumors based on magnetic resonance imaging (MRI) and histopathological imaging (H&E) clinical information. Deep Convolutional Neural Networks (DCNNs) architectures include tens to hundreds of processing layers that can extract multiple levels of features in image-based data, which would be otherwise very difficult and time-consuming to be recognized and extracted by experts for classification of tumors into different tumor types, as well as segmentation of tumor images. This article summarizes the latest studies of deep learning techniques applied to three different kinds of brain cancer medical images (histology, magnetic resonance, and computed tomography) and highlights current challenges in the field for the broader applicability of DCNN in personalized brain cancer care by focusing on two main applications of DCNNs: classification and segmentation of brain cancer tumors images.

Breast Cancer Classification from Histopathological Images with Inception Recurrent Residual Convolutional Neural Network

The Deep Convolutional Neural Network (DCNN) is one of the most powerful and successful deep learning approaches. DCNNs have already provided superior performance in different modalities of medical imaging including breast cancer classification, segmentation, and detection. Breast cancer is one of the most common and dangerous cancers impacting women worldwide. In this paper, we have proposed a method for breast cancer classification with the Inception Recurrent Residual Convolutional Neural Network (IRRCNN) model. The IRRCNN is a powerful DCNN model that combines the strength of the Inception Network (Inception-v4), the Residual Network (ResNet), and the Recurrent Convolutional Neural Network (RCNN). The IRRCNN shows superior performance against equivalent Inception Networks, Residual Networks, and RCNNs for object recognition tasks. In this paper, the IRRCNN approach is applied for breast cancer classification on two publicly available datasets including BreakHis and Breast Cancer (BC) classification challenge 2015. The experimental results are compared against the existing machine learning and deep learning-based approaches with respect to image-based, patch-based, image-level, and patient-level classification. The IRRCNN model provides superior classification performance in terms of sensitivity, area under the curve (AUC), the ROC curve, and global accuracy compared to existing approaches for both datasets.

Parking Slot Detection on Around-View Images Using DCNN

Due to the complex visual environment and incomplete display of parking slots on around-view images, vision-based parking slot detection is a major challenge. Previous studies in this field mostly use the existing models to solve the problem, the steps of which are cumbersome. In this paper, we propose a parking slot detection method that uses directional entrance line regression and classification based on a deep convolutional neural network (DCNN) to make it robust and simple. For parking slots with different shapes and observed from different angles, we represent the parking slot as a directional entrance line. Subsequently, we design a DCNN detector to simultaneously obtain the type, position, length, and direction of the entrance line. After that, the complete parking slot can be easily inferred using the detection results and prior geometric information. To verify our method, we conduct experiments on the public ps2.0 dataset and self-annotated parking slot dataset with 2,135 images. The results show that our method not only outperforms state-of-the-art competitors with a precision rate of 99.68% and a recall rate of 99.41% on the ps2.0 dataset but also performs a satisfying generalization on the self-annotated dataset. Moreover, it achieves a real-time detection speed of 13 ms per frame on Titan Xp. By converting the parking slot into a directional entrance line, the specially designed DCNN detector can quickly and effectively detect various types of parking slots.

Fully-automatic defects classification and restoration for STM images

The Scanning tunneling microscope (STM) is a micro instrument designed for surface morphology with nanometer precision. The restoration of the STM image defects usually needs human judgements and manual positioning because of the diversity of the morphology and the randomness of the defects. This paper provides a new fully-automatic method that combines deep convolutional neural classification network and unique restoration algorithms corresponding to different defects. Aimed at automatically processing compound defects in STM images, the method first predicts what kinds of defects a raw STM image has by a series of parallel binary classification networks, and then decides the process order according to the predicted labels, and finally restores the defects by corresponding global restoration algorithms in order. Experiment results prove the provided method can restore the STM images by self-judging, self-positioning, self-processing without any manual intervention.

Multiplanar analysis for pulmonary nodule classification in CT images using deep convolutional neural network and generative adversarial networks

PURPOSE

Early detection and treatment of lung cancer holds great importance. However, pulmonary-nodule classification using CT images alone is difficult to realize. To address this concern, a method for pulmonary-nodule classification based on a deep convolutional neural network (DCNN) and generative adversarial networks (GAN) has previously been proposed by the authors. In that method, the said classification was performed exclusively using axial cross sections of pulmonary nodules. During actual medical-examination procedures, however, a comprehensive judgment can only be made via observation of various pulmonary-nodule cross sections. In the present study, a comprehensive analysis was performed by extending the application of the previously proposed DCNN- and GAN-based automatic classification method to multiple cross sections of pulmonary nodules.

METHODS

Using the proposed method, CT images of 60 cases with confirmed pathological diagnosis by biopsy are analyzed. Firstly, multiplanar images of the pulmonary nodule are generated. Classification training was performed for three DCNNs. A certain pretraining was initially performed using GAN-generated nodule images. This was followed by fine-tuning of each pretrained DCNN using original nodule images provided as input.

RESULTS

As a result of the evaluation, the specificity was 77.8% and the sensitivity was 93.9%. Additionally, the specificity was observed to have improved by 11.1% without any reduction in the sensitivity, compared to our previous report.

CONCLUSION

This study reports development of a comprehensive analysis method to classify pulmonary nodules at multiple sections using GAN and DCNN. The effectiveness of the proposed discrimination method based on use of multiplanar images has been demonstrated to be improved compared to that realized in a previous study reported by the authors. In addition, the possibility of enhancing classification accuracy via application of GAN-generated images, instead of data augmentation, for pretraining even for medical datasets that contain relatively few images has also been demonstrated.

Prediction of Enzyme Function Based on Three Parallel Deep CNN and Amino Acid Mutation

During the past decade, due to the number of proteins in PDB database being increased gradually, traditional methods cannot better understand the function of newly discovered enzymes in chemical reactions. Computational models and protein feature representation for predicting enzymatic function are more important. Most of existing methods for predicting enzymatic function have used protein geometric structure or protein sequence alone. In this paper, the functions of enzymes are predicted from many-sided biological information including sequence information and structure information. Firstly, we extract the mutation information from amino acids sequence by the position scoring matrix and express structure information with amino acids distance and angle. Then, we use histogram to show the extracted sequence and structural features respectively. Meanwhile, we establish a network model of three parallel Deep Convolutional Neural Networks (DCNN) to learn three features of enzyme for function prediction simultaneously, and the outputs are fused through two different architectures. Finally, The proposed model was investigated on a large dataset of 43,843 enzymes from the PDB and achieved 92.34% correct classification when sequence information is considered, demonstrating an improvement compared with the previous result.

An anti-forensic scheme on computer graphic images and natural images using generative adversarial networks

Computer graphic images (CGI) can be manufactured very similar to natural images (NI) by state-of-the-art algorithms in computer graphic filed. Thus, there are various identification algorithms proposed to detect CGI. However, the manipulation is complicated and difficult for an ultimate CGI against the forensic algorithms. Further, the forensics on CGI and NI made achievements in the different aspects with the encouragement of deep learning. Though the generated CGI can achieve high quality automatically by generative adversarial networks (GAN), CGI generation based on GAN is difficult to ensure that it cannot be detected by forensics. In this paper, we propose a brief and effective architecture based on GAN for preventing the generated images being detected under the forensics on CGI and NI. The adapted characteristics will make the CGI generated by GAN fools the detector and keep the end-to-end generation mode of GAN.