Deep Learning Tools Research Articles

Detection of aspiration from images of a videofluoroscopic swallowing study adopting deep learning.

A videofluoroscopic swallowing study (VFSS) is conducted to detect aspiration. However, aspiration occurs within a short time and is difficult to detect. If deep learning can detect aspirations with high accuracy, clinicians can focus on the diagnosis of the detected aspirations. Whether VFSS aspirations can be classified using rapid-prototyping deep-learning tools was studied. VFSS videos were separated into individual image frames. A region of interest was defined on the pharynx. Three convolutional neural networks (CNNs), namely a Simple-Layer CNN, Multiple-Layer CNN, and Modified LeNet, were designed for the classification. The performance results of the CNNs were compared in terms of the areas under their receiver-operating characteristic curves (AUCs). A total of 18,333 images obtained through data augmentation were selected for the evaluation. The different CNNs yielded sensitivities of 78.8%-87.6%, specificities of 91.9%-98.1%, and overall accuracies of 85.8%-91.7%. The AUC of 0.974 obtained for the Simple-Layer CNN and Modified LeNet was significantly higher than that obtained for the Multiple-Layer CNN (AUC of 0.936) (p < 0.001). The results of this study show that deep learning has potential for detecting aspiration with high accuracy.

A comparative analysis of chronic obstructive pulmonary disease using machine learning, and deep learning

&lt;span lang="EN-US"&gt;Chronic obstructive pulmonary disease (COPD) is a general clinical issue in numerous countries considered the fifth reason for inability and the third reason for mortality on a global scale within 2021. From recent reviews, a deep convolutional neural network (CNN) is used in the primary analysis of the deadly COPD, which uses the computed tomography (CT) images procured from the deep learning tools. Detection and analysis of COPD using several image processing techniques, deep learning models, and machine learning models are notable contributions to this review. This research aims to cover the detailed findings on pulmonary diseases or lung diseases, their causes, and symptoms, which will help treat infections with high performance and a swift response. The articles selected have more than 80% accuracy and are tabulated and analyzed for sensitivity, specificity, and area under the curve (AUC) using different methodologies. This research focuses on the various tools and techniques used in COPD analysis and eventually provides an overview of COPD with coronavirus disease 2019 (COVID-19) symptoms. &lt;/span&gt;

Rapid coarse-grained simulation of whole cryo-tomograms.

Cryo-electron tomography is a rapidly expanding field of microscopy investigation, from subtomogram averaging to cellular tomography. This range of applications leads to highly varied data which can be difficult to analyze systematically. Multiple deep-learning techniques have been successfully applied to automate object classification, aiding in particle averaging or structural analysis. However, these neural networks typically rely on large training datasets that are painstakingly generated by hand based on a subset of the data. A method that does not require detailed hand-segmentation by an expert is needed to allow high-throughput application of deep-learning tools to varied datasets. Here we present CTS, a toolbox for modeling vitreous fields of randomized proteins and other organic structures and simulating these fields as complete tomograms. Crucially, CTS also generates a molecular atlas that provides efficient inputs to deep learning tools. This simple-to-use program can generate a complete simulation and atlas in minutes, and accepts a wide range of structure files to generate independent training data. The simulation replicates imaging processes to produce a realistic level of CTF aberration, resolution, contrast, and noise to best provide high-quality training data that is still representative of target data. CTS can model multiple structural organizations, such as bundled fibers, uniform and non-uniform protein complexes, and particle clusters in addition to modeling carbon gridholes, vitreous ice, and vesicular membranes. These simulations have proven to be capable of training highly accurate deep-learning segmentations of acellular tomograms with no further human input. Complex cellular tomograms are segmented more reliably and with less manual training required when paired with simulated training datasets, in addition to improving accuracy especially with fine-grained structural differences that can be difficult to discern and segment by hand. This tool therefore lowers the barrier for broad and novel deep-learning applications in cryo-ET.

More eyes on the prize: open-source data, software and hardware for advancing plant science through collaboration.

Automating the analysis of plants using image processing would help remove barriers to phenotyping and large-scale precision agricultural technologies, such as site-specific weed control. The combination of accessible hardware and high-performance deep learning (DL) tools for plant analysis is becoming widely recognised as a path forward for both plant science and applied precision agricultural purposes. Yet, a lack of collaboration in image analysis for plant science, despite the open-source origins of much of the technology, is hindering development. Here, we show how tools developed for specific attributes of phenotyping or weed recognition for precision weed control have substantial overlapping data structure, software/hardware requirements and outputs. An open-source approach to these tools facilitates interdisciplinary collaboration, avoiding unnecessary repetition and allowing research groups in both basic and applied sciences to capitalise on advancements and resolve respective bottlenecks. The approach mimics that of machine learning in its nascence. Three areas of collaboration are identified as critical for improving efficiency, (1) standardized, open-source, annotated dataset development with consistent metadata reporting; (2) establishment of accessible and reliable training and testing platforms for DL algorithms; and (3) sharing of all source code used in the research process. The complexity of imaging plants and cost of annotating image datasets means that collaboration from typically distinct fields will be necessary to capitalize on the benefits of DL for both applied and basic science purposes.

Physical Quantities Reconstruction in Reacting Flows with Deep Learning

Performing measurements in reacting flows is a challenging task due to the complexity of measuring all quantities of interest simultaneously or limitations in the optical access. To compensate for this, recent advances in deep learning have shown a strong potential in augmenting the information content in datasets composed of partial measurements by reconstructing the quantities that could not be measured. The present work analyses the use of such deep learning tools for the reconstruction of quantities in two different cases. First, Convolutional Neural Networks (CNNs) are used to reconstruct the heat release rate (HRR) from velocity measurements in a methane/air premixed flame under harmonic excitation. The CNNs are trained from complete datasets at some specific frequencies and amplitudes of excitation and their ablility to reconstruct the HRR for different operating conditions with good accuracy is demonstrated. Secondly, an alternate approach based on Physics-Informed Neural Networks that do not require the training data to have all the quantities is explored. It is applied to a puffing pool fire where the velocity field is reconstructed from observations of pressure, temperature and density with good accuracy. Both approaches are also shown to be robust with respect to noise.

Car Detection from Very High-Resolution UAV Images Using Deep Learning Algorithms

It is important to determine car density in parking lots, especially in hospitals, large enterprises, and residential areas, which are used intensively, in terms of executing existing management systems and making precise plans for the future. In this study, cars in parking lots were detected using high-resolution unmanned aerial vehicle (UAV) images with deep learning methods. We tested the performance of the two approaches by determining the number of cars in a parking lot using the You Only Look Once (YOLOv3) and Mask Region–Based Convolutional Neural Networks (Mask R-CNN) approaches as deep learning methods and the deep learning tool of Esri ArcGIS Pro. High-resolution UAV images were processed by photogrammetry and used as input products for the R-CNN and YOLOv3 algorithm. Recall, F1 score, precision ratio/uncertainty accuracy, and average producer accuracy of products automatically extracted with the algorithm were determined as 0.862/0.941, 0.874/0.946, 0.885/0.951, and 0.776/0.897 for R-CNN and YOLOv3, respectively.

Deep Learning Identifies HAT1 as a Morphological Regulator in Esophageal Squamous Carcinoma Cells through Controlling Cell Senescence

Histopathology is a critical approach for diagnostic tasks and precision treatment. However, histopathological deep learning tools for auto‐identification remain poorly developed. Meanwhile, the interpretation of the computer vision attention into a cellular process is less efficient in a systematic way. Herein, it is identified that histone acetyltransferase 1 (HAT1) is an aging‐associated gene in the esophagus epithelium by machine learning. An interpretable deep learning model is developed to distinguish morphological changes with varied HAT1 expressions in esophageal squamous carcinoma cells (ESCC). The gradient‐weighted class activation mapping and prediction score analysis reveal that the computer's vision focuses on the nuclear sizes of ESCC. The hypothesized phenotype is verified in HAT1‐knockdown ESCCs. Finally, HAT1 regulating cell senescence by affecting the H3K27 acetylation and E2F transcription factor 7 (E2F7) expression is shown. Herein, the feasibility and benefits of applying histopathological deep learning assistance systems in routine practice scenarios and connecting phenotype and genotype for further genetic research are suggested.

Deep neural network architecture for automated soft surgical skills evaluation using objective structured assessment of technical skills criteria.

Classic methods of surgery skills evaluation tend to classify the surgeon performance in multi-categorical discrete classes. If this classification scheme has proven to be effective, it does not provide in-between evaluation levels. If these intermediate scoring levels were available, they would provide more accurate evaluation of the surgeon trainee. We propose a novel approach to assess surgery skills on a continuous scale ranging from 1 to 5. We show that the proposed approach is flexible enough to be used either for scores of global performance or several sub-scores based on a surgical criteria set called Objective Structured Assessment of Technical Skills (OSATS). We established a combined CNN+BiLSTM architecture to take advantage of both temporal and spatial features of kinematic data. Our experimental validation relies on real-world data obtained from JIGSAWS database. The surgeons are evaluated on three tasks: Knot-Tying, Needle-Passing and Suturing. The proposed framework of neural networks takes as inputs a sequence of 76 kinematic variables and produces an output float score ranging from 1 to 5, reflecting the quality of the performed surgical task. Our proposed model achieves high-quality OSATS scores predictions with means of Spearman correlation coefficients between the predicted outputs and the ground-truth outputs of 0.82, 0.60 and 0.65 for Knot-Tying, Needle-Passing and Suturing, respectively. To our knowledge, we are the first to achieve this regression performance using the OSATS criteria and the JIGSAWS kinematic data. An effective deep learning tool was created for the purpose of surgical skills assessment. It was shown that our method could be a promising surgical skills evaluation tool for surgical training programs.

Automatic treatment outcome prediction with DeepInteg based on multimodal radiological images in rectal cancer

Neoadjuvant systemic treatment before surgery is a prevalent regimen in the patients with advanced-stage or high-risk tumor, which has shaped the treatment strategies and cancer survival in the past decades. However, some patients present with poor response to the neoadjuvant treatment. Therefore, it is of great significance to develop tools to help distinguish the patients that could achieve pathological complete response before surgery to avoid inappropriate treatment. Here, this study demonstrated a multi-task deep learning tool called DeepInteg. In the DeepInteg framework, the segmentation module was constructed based on the CE-Net with a context extractor to achieve end-to-end delineation of region of interest (ROI) from radiological images, then the features of segmented Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) images of each case were fused and input to the classification module based on a convolution neural network for treatment outcome prediction. The dataset with 1700 MRI and CT slices collected from the prospectively randomized clinical trial (NCT01211210) on systemic treatment for rectal cancer was used to develop and systematically optimize DeepInteg. As a result, DeepInteg achieved automatic segmentation of tumoral ROI with Dices of 0.766 and 0.719 and mIoUs of 0.788 and 0.756 in CT and MRI images, respectively. In addition, DeepInteg achieved AUC of 0.833, accuracy of 0.826 and specificity of 0.856 in the prediction for pathological complete response after treatment, which showed better performance compared with the model based on CT or MRI alone. This study provide a robust framework to develop disease-specific tools for automatic delineation of ROI and clinical outcome prediction. The well-trained DeepInteg could be readily applied in clinic to predict pathological complete response after neoadjuvant therapy in rectal cancer patients.

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.

Forecasting commodity prices: empirical evidence using deep learning tools.

Since the last two decades, financial markets have exhibited several transformations owing to recurring crises episodes that has led to the development of alternative assets. Particularly, the commodity market has attracted attention from investors and hedgers. However, the operational research stream has also developed substantially based on the growth of the artificial intelligence field, which includes machine learning and deep learning. The choice of algorithms in both machine learning and deep learning is case-sensitive. Hence, AI practitioners should first attempt solutions related to machine learning algorithms, and if such solutions are unsatisfactory, they must apply deep learning algorithms. Using this perspective, this study aims to investigate the potential of various deep learning basic algorithms for forecasting selected commodity prices. Formally, we use the Bloomberg Commodity Index (noted by the Global Aggregate Index) and its five component indices: Bloomberg Agriculture Subindex, Bloomberg Precious Metals Subindex, Bloomberg Livestock Subindex, Bloomberg Industrial Metals Subindex, and Bloomberg Energy Subindex. Based on daily data from January 2002 (the beginning wave of commodity markets' financialization) to December 2020, results show the effectiveness of the Long Short-Term Memory method as a forecasting tool and the superiority of the Bloomberg Livestock Subindex and Bloomberg Industrial Metals Subindex for assessing other commodities' indices. These findings is important in term for investors in term of risk management as well as policymakers in adjusting public policy, especially during Russian-Ukrainian war.

A Novel Multi-Objective Binary Chimp Optimization Algorithm for Optimal Feature Selection: Application of Deep-Learning-Based Approaches for SAR Image Classification.

Removing redundant features and improving classifier performance necessitates the use of meta-heuristic and deep learning (DL) algorithms in feature selection and classification problems. With the maturity of DL tools, many data-driven polarimetric synthetic aperture radar (POLSAR) representation models have been suggested, most of which are based on deep convolutional neural networks (DCNNs). In this paper, we propose a hybrid approach of a new multi-objective binary chimp optimization algorithm (MOBChOA) and DCNN for optimal feature selection. We implemented the proposed method to classify POLSAR images from San Francisco, USA. To do so, we first performed the necessary preprocessing, including speckle reduction, radiometric calibration, and feature extraction. After that, we implemented the proposed MOBChOA for optimal feature selection. Finally, we trained the fully connected DCNN to classify the pixels into specific land-cover labels. We evaluated the performance of the proposed MOBChOA-DCNN in comparison with nine competitive methods. Our experimental results with the POLSAR image datasets show that the proposed architecture had a great performance for different important optimization parameters. The proposed MOBChOA-DCNN provided fewer features (27) and the highest overall accuracy. The overall accuracy values of MOBChOA-DCNN on the training and validation datasets were 96.89% and 96.13%, respectively, which were the best results. The overall accuracy of SVM was 89.30%, which was the worst result. The results of the proposed MOBChOA on two real-world benchmark problems were also better than the results with the other methods. Furthermore, it was shown that the MOBChOA-DCNN performed better than methods from previous studies.

Using Multilingual Bidirectional Encoder Representations from Transformers on Medical Corpus for Kurdish Text Classification

Technology has dominated a huge part of human life. Furthermore, technology users use language continuously to express feelings and sentiments about things. The science behind identifying human attitudes toward a particular product, service,or topic is one of the most active fields of research, and it is called sentiment analysis. While the English language is making real progress in sentiment analysis daily, other less-resourced languages, such as Kurdish, still suffer from fundamental issues and challenges in Natural Language Processing (NLP). This paper experimentswith the recently published medical corpus using the classical machine learning method and the latest deep learning tool in NLP and Bidirectional Encoder Representations from Transformers (BERT). We evaluated the findings of both machine learning and deep learning. The outcome indicates that BERT outperforms all the machine learning classifiers by scoring (92%) in accuracy, which is by two points higher than machine learning classifiers.

Cross-task cognitive workload recognition using a dynamic residual network with attention mechanism based on neurophysiological signals

Background and ObjectiveEvaluation of human cognitive workload (CW) helps improve the user experience of human-centered systems. To provide a continuous estimation of the CW, we built a CW recognizer that maps human electroencephalograms (EEGs) to discrete CW levels with deep learning tools. However, the EEG distribution varies when humans perform different cognitive tasks. There is thus a question on the capacity for generalizing the CW recognizer across tasks. In this study, we examined the CW's performance when it was trained and tested on two EEG databases corresponding to different human-machine tasks. MethodsA novel deep neural network-based EEG recognizer, dynamic residual network with attention mechanism (DRNA-Net), is proposed in the present study. By taking advantage of recurrent networks, the DRNA-Net further incorporates a self-attention mechanism in discovering robust EEG patterns across different cognitive tasks. ResultsWe designed an experiment that applied a multidimensional N-back task to induce the CW that consists of visual and auditory memory tasks. We validated the cross-task generalization capability of the DRNA-Net based on the EEG features extracted from the N-back task and a public database. The results show that the DRNA-Net achieves classification accuracy and Macro-F1 values are 0.6055 and 0.6067, respectively. ConclusionsThe performance of the DRNA-Net indicates that it has a certain ability of cross-task cognitive workload classification, which outperforms several shallow learners and deep convolutional neural networks under various conditions of the feature subsets.

Stochastic Study on CNN Approach for Classifying Images

Recent years have seen a surge in interest in artificial intelligence and computer vision applications, which are essential for many real-time tasks like video summarization, picture retrieval, and image categorization. Convolutional neural networks, which are employed in numerous image processing and computer vision applications, are one of the most popular deep learning techniques. In this study, a CNN model for classifying colour images is proposed. The proposed model was developed using deep learning tools in MATLAB. Additionally, three separate datasets of varying sizes were used to test the suggested model. With the largest dataset, the suggested model produced the highest accuracy, precision, and sensitivity results, which are as follows: accuracy is 0.9924, precision is 0.9947, and sensitivity is 0.9931 when compared to other models

Artificial intelligence applied in neoantigen identification facilitates personalized cancer immunotherapy.

The field of cancer neoantigen investigation has developed swiftly in the past decade. Predicting novel and true neoantigens derived from large multi-omics data became difficult but critical challenges. The rise of Artificial Intelligence (AI) or Machine Learning (ML) in biomedicine application has brought benefits to strengthen the current computational pipeline for neoantigen prediction. ML algorithms offer powerful tools to recognize the multidimensional nature of the omics data and therefore extract the key neoantigen features enabling a successful discovery of new neoantigens. The present review aims to outline the significant technology progress of machine learning approaches, especially the newly deep learning tools and pipelines, that were recently applied in neoantigen prediction. In this review article, we summarize the current state-of-the-art tools developed to predict neoantigens. The standard workflow includes calling genetic variants in paired tumor and blood samples, and rating the binding affinity between mutated peptide, MHC (I and II) and T cell receptor (TCR), followed by characterizing the immunogenicity of tumor epitopes. More specifically, we highlight the outstanding feature extraction tools and multi-layer neural network architectures in typical ML models. It is noted that more integrated neoantigen-predicting pipelines are constructed with hybrid or combined ML algorithms instead of conventional machine learning models. In addition, the trends and challenges in further optimizing and integrating the existing pipelines are discussed.

A combined variational encoding and optimization framework for design of the water–energy–food nexus

Nowadays, water, energy and food supplies are among the main problems faced by humanity. These issues are strongly linked and must be addressed by considering their interactions. The linkage is formulated as a nonlinear multiobjective optimization problem. Moreover, the solution can benefit from large databases that lead to the development of surrogate models. In this work, surrogate models are used to incorporate recorded input information. To avoid large-scale models, we use generative deep learning strategies to build low-order models. The surrogate models are coupled to the optimization problem to render optimal operation and structure of the nexus. The results show that the reduced model benefits the performance of the optimization problem. Furthermore, we implemented a clustering algorithm to reduce the number of solutions found by the multiobjective approach. The proposed methodology establishes a contribution for the integration of traditional optimization and deep learning tools to address the complex interactions of the water–energy–food nexus.

A comparison of visual place recognition methods using a mobile robot in an indoor environment

Spatial awareness is an important competence for a mobile robotic system. A robot needs to localise and perform context interpretation to provide any meaningful service. With the deep learning tools and readily available sensors, visual place recognition is a first step towards identifying the environment to bring a robot closer to spatial awareness. In this paper, we implement place recognition on a mobile robot considering a deep learning approach. For simple place classification, where the task involves classifying images into a limited number of categories, all three architectures; VGG16, Inception-v3 and ResNet50, perform well. However, considering the pros and cons, the choice may depend on available computational resources and deployment constraints.

LSTM Based 24 hours Ahead Forecasting of Solar PV System for Standalone Household System

A rapid expansion in the technology of solar PV energy in recent years has paved the way for PV market to grow resulting in a cost reduction in material. Hence, technological improvements are eminent to maintain the stable working of the system. In order to create a system many techniques and modules are used like, MPPT algorithms to track the power which increases real-time efficiency, and a battery management system to efficiently manage the stored energy from the battery. This paper aims to design a forecasting model to predict the weather and load of the standalone system to prepare the battery for future use. The forecasting is done using Machine learning and Deep learning tools like RNN, LSTM, and GRU and the results are fed into the system in MATLAB Simulink simulation platform.

Object identification and surveillance based on deep learning algorithms for quadcopters

Drone technology is evolving for the applications like surveillance, observation, rescue and control crimes, military, agriculture, civil and many more purposes. But these surveillance systems are monitored by human interaction so there may be some negligence and malfunction may happen due to lack of observation. The objective of the work was development of unmanned aerial vehicle and implementation of the deep learning, image processing tools in the quad-copter based surveillance system. Development of the quad-copter was done by using necessary components such as Arms and power distribution board with help of fasteners. And the brushless dc motors are fitted with electronic speed controllers. And the suitable propellers are fitted to the motors, from which thrust is obtained. The Arduino Uno microcontroller is used as flight controller with MPU6050. Sensor fusion concept coding was used for accelerometer and gyroscope for stable direction orientations of Aerial vehicle. Multiwii platform was used to build the flight controller for achieving desire rotation of motors as well as proper directions and speed. The receiver was installed in the quad-copter for wireless control with transmitter of 2.4 GHz range. And IP camera was used, from which the surveillance visuals are taken for monitoring. Battery of 2200mah capacity of 3 cells was used for power supply of whole system. The visuals were obtained in raspberry pi, the live video stream/images are processed with the deep learning tools i.e., Open CV, Tensor flow, yolo for effective surveillance.