DL-based Techniques Research Articles

Hybrid deep features computed from spatial images and bit plane-based pattern maps for the classification of chest X-ray images

Chest X-ray images are known to be extremely helpful in the investigation of a numerous pulmonary conditions such as COVID-19 and pneumonia. Many affected individuals may be protected from such pulmonary conditions through early identification. Unfortunately, COVID-19 can be misdiagnosed as pneumonia, which can rapidly worsen and lead to death. The sensitivity of RT-PCR-based COVID-19 detection is also not satisfactory. Herein, a deep-learning (DL) model for predicting three distinct classes, i.e., COVID-19, pneumonia, and normal, is presented, which achieves good classification performance. Three separate publicly available datasets and an additional dataset with their merged form were used to confirm the efficacy of the proposed model. The DL-based techniques compute features from original raw input spatial images and do not directly provide much information on extremely fine image details, which is quite important for biomedical image analysis. As the individual image bit planes (BPs) carry extremely-fine-to-coarse image information and the effective handcrafted pattern maps created from these BPs may incorporate important discriminating information, the deep features computed from such handcrafted pattern maps may provide complementary information regarding the deep features computed using raw spatial input images. Therefore, we propose a blend of deep features computed with raw spatial images and deep features computed using the proposed local bit plane-based pattern maps to predict the three classes. It is demonstrated that the blend of such features supplies improved discrimination potential and is complementary to the sole features. We incorporated multiscale information computed in each BPs along with interscale details to generate the final bit plane-based pattern maps. The proposed model achieved an average accuracy of 100%, 99.9%, 98.8%, and 98.8% for datasets 1,2, and 3 and their combined form, respectively, outperforming the existing methods.

Classification of Cardiovascular Arrhythmia Using Deep Learning Techniques: A Review

Deep Learning (DL), an offshoot of Machine Learning (ML) has emerged as a powerful and feasible solution for medical image analysis due to advancements in robust computer software and hardware technologies. It plays a key role in Cardiovascular disease (CVD) diagnosis by detecting anomalies in Electrocardiogram (ECG) signals. Cardiac arrhythmia, which refers to irregular heartbeat, may signal an early symptom of CVD and can lead to fatal outcomes if ignored. Accurate detection of arrhythmia is very challenging even for experts to distinguish between acute and chronic conditions in ECG readings. This triggered the focus of researchers to explore the application of Artificial Intelligence for ECG classification. Traditional machine learning methods use handcrafted features that require domain knowledge. The new era in DL makes the automatic detection of Cardiovascular Disease (CVD) possible. In this paper, an exhaustive review of DL-based techniques for ECG classification has been presented. Research findings in this survey indicate the challenges and issues with arrhythmia detection, such as single lead and multiple lead ECG signals, choice of the size of the training data set, and the number of arrhythmia classes, etc. The study also signifies that there is great scope for improving the performance of arrhythmia prediction models by employing hybrid ensemble learning, time series analysis using Recurrent Neural Network architectures and identification of unexplored classes of arrhythmia.

Evaluation of High-Dimensional Data Classification for Skin Malignancy Detection Using DL-Based Techniques

Skin cancer can be detected through visual screening and skin analysis based on the biopsy and pathological state of the human body. The survival rate of cancer patients is low, and millions of people are diagnosed annually. By determining the different comparative analyses, the skin malignancy classification is evaluated. Using the Isomap with the vision transformer, we analyze the high-dimensional images with dimensionality reduction. Skin cancer can present with severe cases and life-threatening symptoms. Overall performance evaluation and classification tend to improve the accuracy of the high-dimensional skin lesion dataset when completed. In deep learning methodologies, the distinct phases of skin malignancy classification are determined by its accuracy, specificity, F1 recall, and sensitivity while implementing the classification methodology. A nonlinear dimensionality reduction technique called Isomap preserves the data’s underlying nonlinear relationships intact. This is essential for the categorization of skin malignancies, as the features that separate malignant from benign skin lesions may not be linearly separable. Isomap decreases the data’s complexity while maintaining its essential characteristics, making it simpler to analyze and explain the findings. High-dimensional datasets for skin lesions have been evaluated and classified more effectively when evaluated and classified using Isomap with the vision transformer.

Basal Cell Carcinoma Diagnosis with Fusion of Deep Learning and Telangiectasia Features.

In recent years, deep learning (DL) has been used extensively and successfully to diagnose different cancers in dermoscopic images. However, most approaches lack clinical inputs supported by dermatologists that could aid in higher accuracy and explainability. To dermatologists, the presence of telangiectasia, or narrow blood vessels that typically appear serpiginous or arborizing, is a critical indicator of basal cell carcinoma (BCC). Exploiting the feature information present in telangiectasia through a combination of DL-based techniques could create a pathway for both, improving DL results as well as aiding dermatologists in BCC diagnosis. This study demonstrates a novel "fusion" technique for BCC vs non-BCC classification using ensemble learning on a combination of (a) handcrafted features from semantically segmented telangiectasia (U-Net-based) and (b) deep learning features generated from whole lesion images (EfficientNet-B5-based). This fusion method achieves a binary classification accuracy of 97.2%, with a 1.3% improvement over the corresponding DL-only model, on a holdout test set of 395 images. An increase of 3.7% in sensitivity, 1.5% in specificity, and 1.5% in precision along with an AUC of 0.99 was also achieved. Metric improvements were demonstrated in three stages: (1) the addition of handcrafted telangiectasia features to deep learning features, (2) including areas near telangiectasia (surround areas), (3) discarding the noisy lower-importance features through feature importance. Another novel approach to feature finding with weak annotations through the examination of the surrounding areas of telangiectasia is offered in this study. The experimental results show state-of-the-art accuracy and precision in the diagnosis of BCC, compared to three benchmark techniques. Further exploration of deep learning techniques for individual dermoscopy feature detection is warranted.

Try with Simpler – An Evaluation of Improved Principal Component Analysis in Log-based Anomaly Detection

With the rapid development of deep learning (DL), the recent trend of log-based anomaly detection focuses on extracting semantic information from log events (i.e., templates of log messages) and designing more advanced DL models for anomaly detection. Indeed, the effectiveness of log-based anomaly detection can be improved, but these DL-based techniques further suffer from the limitations of more heavy dependency on training data (such as data quality or data labels) and higher costs in time and resources due to the complexity and scale of DL models, which hinder their practical use. On the contrary, the techniques based on traditional machine learning or data mining algorithms are less dependent on training data and more efficient, but produce worse effectiveness than DL-based techniques which is mainly caused by the problem of unseen log events (some log events in incoming log messages are unseen in training data) confirmed by our motivating study. Intuitively, if we can improve the effectiveness of traditional techniques to be comparable with advanced DL-based techniques, log-based anomaly detection can be more practical. Indeed, an existing study in the other area (i.e., linking questions posted on Stack Overflow) has pointed out that traditional techniques with some optimizations can indeed achieve comparable effectiveness with the state-of-the-art DL-based technique, indicating the feasibility of enhancing traditional log-based anomaly detection techniques to some degree. Inspired by the idea of “try-with-simpler”, we conducted the first empirical study to explore the potential of improving traditional techniques for more practical log-based anomaly detection. In this work, we optimized the traditional unsupervised PCA (Principal Component Analysis) technique by incorporating a lightweight semantic-based log representation in it, called SemPCA , and conducted an extensive study to investigate the potential of SemPCA for more practical log-based anomaly detection. By comparing seven log-based anomaly detection techniques (including four DL-based techniques, two traditional techniques, and SemPCA ) on both public and industrial datasets, our results show that SemPCA achieves comparable effectiveness as advanced supervised/semi-supervised DL-based techniques while being much more stable under insufficient training data and more efficient, demonstrating that the traditional technique can still excel after small but useful adaptation.

Securing the Dynamic Realm: A Comprehensive Review of ML Algorithms in IoT-Based Home Automation Systems and Beyond

This paper comprehensively reviews the imperative for secure IoT systems, emphasizing the challenges posed by their dynamic nature. Exploring various ML algorithms for IoT security, it highlights their advantages while addressing common limitations, including computational overhead and privacy risks. The focus narrows to federated learning (FL) and deep learning (DL) algorithms, showcasing their potential to overcome conventional ML drawbacks by preserving data privacy. The study provides an in-depth analysis of FL and DL-based techniques, emphasizing their efficiency in enhancing security in IoT-based home automation systems. The paper further examines ML's pivotal role in smart homes, presenting a case study that utilizes the support vector machine algorithm to distinguish between regular occupants and intruders. Extending the discussion to face recognition for home automation, the review underscores the utilization of IoT and smart techniques. Beyond home automation, the paper delves into the broader landscape of ML applications in the Fourth Industrial Revolution, offering insights into cybersecurity, smart cities, healthcare, and more. The review briefly introduces the utilization of Convolutional Neural Networks (CNNs) within the broader context of deep learning algorithms. While the main emphasis remains on FL and DL, the paper acknowledges CNNs as a powerful tool for image-based tasks, especially relevant in the context of visual data analysis for security in IoT-based home automation systems. In summary, this concise review encapsulates the transformative impact of ML on IoT-based home automation security, providing valuable perspectives on current trends, challenges, and future research directions. The inclusion of CNNs within the abstract recognizes their relevance, especially in image-based security applications.

Android Ransomware Detection Using Supervised Machine Learning Techniques Based on Traffic Analysis.

In today's digitalized era, the usage of Android devices is being extensively witnessed in various sectors. Cybercriminals inevitably adapt to new security technologies and utilize these platforms to exploit vulnerabilities for nefarious purposes, such as stealing users' sensitive and personal data. This may result in financial losses, discredit, ransomware, or the spreading of infectious malware and other catastrophic cyber-attacks. Due to the fact that ransomware encrypts user data and requests a ransom payment in exchange for the decryption key, it is one of the most devastating types of malicious software. The implications of ransomware attacks can range from a loss of essential data to a disruption of business operations and significant monetary damage. Artificial intelligence (AI)-based techniques, namely machine learning (ML), have proven to be notable in the detection of Android ransomware attacks. However, ensemble models and deep learning (DL) models have not been sufficiently explored. Therefore, in this study, we utilized ML- and DL-based techniques to build efficient, precise, and robust models for binary classification. A publicly available dataset from Kaggle consisting of 392,035 records with benign traffic and 10 different types of Android ransomware attacks was used to train and test the models. Two experiments were carried out. In experiment 1, all the features of the dataset were used. In experiment 2, only the best 19 features were used. The deployed models included a decision tree (DT), support vector machine (SVM), k-nearest neighbor (KNN), ensemble of (DT, SVM, and KNN), feedforward neural network (FNN), and tabular attention network (TabNet). Overall, the experiments yielded excellent results. DT outperformed the others, with an accuracy of 97.24%, precision of 98.50%, and F1-score of 98.45%. Whereas, in terms of the highest recall, SVM achieved 100%. The acquired results were thoroughly discussed, in addition to addressing limitations and exploring potential directions for future work.

A Multivariate Time Series Analysis of Electrical Load Forecasting Based on a Hybrid Feature Selection Approach and Explainable Deep Learning

In the smart grid paradigm, precise electrical load forecasting (ELF) offers significant advantages for enhancing grid reliability and informing energy planning decisions. Specifically, mid-term ELF is a key priority for power system planning and operation. Although statistical methods were primarily used because ELF is a time series problem, deep learning (DL)-based forecasting approaches are more commonly employed and successful in achieving precise predictions. However, these DL-based techniques, known as black box models, lack interpretability. When interpreting the DL model, employing explainable artificial intelligence (XAI) yields significant advantages by extracting meaningful information from the DL model outputs and the causal relationships among various factors. On the contrary, precise load forecasting necessitates employing feature engineering to identify pertinent input features and determine optimal time lags. This research study strives to accomplish a mid-term forecast of ELF study load utilizing aggregated electrical load consumption data, while considering the aforementioned critical aspects. A hybrid framework for feature selection and extraction is proposed for electric load forecasting. Technical term abbreviations are explained upon first use. The feature selection phase employs a combination of filter, Pearson correlation (PC), embedded random forest regressor (RFR) and decision tree regressor (DTR) methods to determine the correlation and significance of each feature. In the feature extraction phase, we utilized a wrapper-based technique called recursive feature elimination cross-validation (RFECV) to eliminate redundant features. Multi-step-ahead time series forecasting is conducted utilizing three distinct long-short term memory (LSTM) models: basic LSTM, bi-directional LSTM (Bi-LSTM) and attention-based LSTM models to accurately predict electrical load consumption thirty days in advance. Through numerous studies, a reduction in forecasting errors of nearly 50% has been attained. Additionally, the local interpretable model-agnostic explanations (LIME) methodology, which is an explainable artificial intelligence (XAI) technique, is utilized for explaining the mid-term ELF model. As far as the authors are aware, XAI has not yet been implemented in mid-term aggregated energy forecasting studies utilizing the ELF method. Quantitative and detailed evaluations have been conducted, with the experimental results indicating that this comprehensive approach is entirely successful in forecasting multivariate mid-term loads.

Is deep learning superior to traditional techniques in machine health monitoring applications

AbstractIn recent years, there has been significant momentum in applying deep learning (DL) to machine health monitoring (MHM). It has been widely claimed that DL methodologies are superior to more traditional techniques in this area. This paper aims to investigate this claim by analysing a real-world dataset of helicopter sensor faults provided by Airbus. Specifically, we will address the problem of machine sensor health unsupervised classification. In a 2019 worldwide competition hosted by Airbus, Fujitsu Systems Europe (FSE) won first prize by achieving an F1-score of 93% using a DL model based on generative adversarial networks (GAN). In another comprehensive study, various modified and existing image encoding methods were compared for the convolutional auto-encoder (CAE) model. The best classification result was achieved using the scalogram as the image encoding method, with an F1-score of 91%. In this paper, we use these two studies as benchmarks to compare with basic statistical analysis methods and the one-class supporting vector machine (SVM). Our comparative study demonstrates that while DL-based techniques have great potential, they are not always superior to traditional methods. We therefore recommend that all future published studies of applying DL methods to MHM include appropriately selected traditional reference methods, wherever possible.

A comprehensive analysis of recent advancements in cancer detection using machine learning and deep learning models for improved diagnostics.

There are millions of people who lose their life due to several types of fatal diseases. Cancer is one of the most fatal diseases which may be due to obesity, alcohol consumption, infections, ultraviolet radiation, smoking, and unhealthy lifestyles. Cancer is abnormal and uncontrolled tissue growth inside the body which may be spread to other body parts other than where it has originated. Hence it is very much required to diagnose the cancer at an early stage to provide correct and timely treatment. Also, manual diagnosis and diagnostic error may cause of the death of many patients hence much research are going on for the automatic and accurate detection of cancer at early stage. In this paper, we have done the comparative analysis of the diagnosis and recent advancement for the detection of various cancer types using traditional machine learning (ML) and deep learning (DL) models. In this study, we have included four types of cancers, brain, lung, skin, and breast and their detection using ML and DL techniques. In extensive review we have included a total of 130 pieces of literature among which 56 are of ML-based and 74 are from DL-based cancer detection techniques. Only the peer reviewed research papers published in the recent 5-year span (2018-2023) have been included for the analysis based on the parameters, year of publication, feature utilized, best model, dataset/images utilized, and best accuracy. We have reviewed ML and DL-based techniques for cancer detection separately and included accuracy as the performance evaluation metrics to maintain the homogeneity while verifying the classifier efficiency. Among all the reviewed literatures, DL techniques achieved the highest accuracy of 100%, while ML techniques achieved 99.89%. The lowest accuracy achieved using DL and ML approaches were 70% and 75.48%, respectively. The difference in accuracy between the highest and lowest performing models is about 28.8% for skin cancer detection. In addition, the key findings, and challenges for each type of cancer detection using ML and DL techniques have been presented. The comparative analysis between the best performing and worst performing models, along with overall key findings and challenges, has been provided for future research purposes. Although the analysis is based on accuracy as the performance metric and various parameters, the results demonstrate a significant scope for improvement in classification efficiency. The paper concludes that both ML and DL techniques hold promise in the early detection of various cancer types. However, the study identifies specific challenges that need to be addressed for the widespread implementation of these techniques in clinical settings. The presented results offer valuable guidance for future research in cancer detection, emphasizing the need for continued advancements in ML and DL-based approaches to improve diagnostic accuracy and ultimately save more lives.

Prediction of ultimate tensile strength of friction stir welding joint using deep learning-based-multilayer perceptron and long short term memory networks

Friction stir welding (FSW) is a solid-state joining technique where the joint strength is mainly influenced by three process parameters, namely, spindle speed (N), welding speed (V), and plunge force (Fz). The modelling of complex relationships between the process parameters and joint strength requires many experiments, which is a challenging, time-consuming, and non-economical affair. To tackle this problem, computational mathematical models such as deep learning (DL) can be employed to predict the joint strength reliably. In this paper, DL techniques, namely, deep multilayer perceptron (DMLP) and long short-term memory (LSTM) networks have been proposed for such a purpose. The DL networks were first trained with the FSW experimental data and then, the pre-trained models were used for predicting the weld strength. It was found that the DMLP and LSTM models provided lower prediction errors, which are RMSE of 3.30 and 7.63, respectively, and can be effectively utilized for determining weld quality. The proposed DL-based techniques were further compared with the traditional models – the shallow artificial neural network (SANN) model having an RMSE of 27.11 and the ANFIS model having an RMSE of 5.31. DMLP was found to be superior in determining the weld strength most accurately.

Deep learning-based registration of two-dimensional dental images with edge specific loss.

Image registration is a very common procedure in dental applications for aligning images. Registration between pairs of images taken from different angles can improve diagnosis. Our study presents an edge-enhanced unsupervised deep learning (DL)-based deformable registration framework for aligning two-dimensional (2D) pairs of dental x-ray images. The proposed neural network is based on the combination of a U-Net like structure, which produces a displacement field, combined with spatial transformer networks, which produce the transformed image. The proposed structure is trained end-to-end by minimizing a weighted loss function consisting of three parts corresponding to image similarity, edge similarity, and registration restrictions. In this regard, the proposed edge specific loss enhances the unsupervised training of the registration framework without the need of supervision through anatomical structures. The proposed framework was applied to two datasets, a set of 104 x-ray images of mandibles, arranged in 2600 pairs for training and testing and a set of 17 pairs of pre- and post-operative reconstructed panoramic images. The proposed model outperformed both conventional registration methods and DL-based techniques for both qualitative and quantitative assessment, in most of the compared metrics concerning intensity similarity and edge distances. The proposed framework achieved accurate and fast deformable alignment of pairs of 2D dental radiographic images. The edge-based module of the loss function enhances the unsupervised learning by directing the network toward deformations that take into consideration the edges of the depicted objects (teeth, bone, and tissue), which are crucial in diagnosis.

Deep learning techniques for detection and prediction of pandemic diseases: a systematic literature review

Deep learning (DL) is becoming a fast-growing field in the medical domain and it helps in the timely detection of any infectious disease (IDs) and is essential to the management of diseases and the prediction of future occurrences. Many scientists and scholars have implemented DL techniques for the detection and prediction of pandemics, IDs and other healthcare-related purposes, these outcomes are with various limitations and research gaps. For the purpose of achieving an accurate, efficient and less complicated DL-based system for the detection and prediction of pandemics, therefore, this study carried out a systematic literature review (SLR) on the detection and prediction of pandemics using DL techniques. The survey is anchored by four objectives and a state-of-the-art review of forty-five papers out of seven hundred and ninety papers retrieved from different scholarly databases was carried out in this study to analyze and evaluate the trend of DL techniques application areas in the detection and prediction of pandemics. This study used various tables and graphs to analyze the extracted related articles from various online scholarly repositories and the analysis showed that DL techniques have a good tool in pandemic detection and prediction. Scopus and Web of Science repositories are given attention in this current because they contain suitable scientific findings in the subject area. Finally, the state-of-the-art review presents forty-four (44) studies of various DL technique performances. The challenges identified from the literature include the low performance of the model due to computational complexities, improper labeling and the absence of a high-quality dataset among others. This survey suggests possible solutions such as the development of improved DL-based techniques or the reduction of the output layer of DL-based architecture for the detection and prediction of pandemic-prone diseases as future considerations.

Contour-guided deep learning based deformable image registration for dose monitoring during CBCT-guided radiotherapy of prostate cancer.

To evaluate deep learning (DL)-based deformable image registration (DIR) for dose accumulation during radiotherapy of prostate cancer patients. Data including 341 CBCTs (209 daily, 132 weekly) and 23 planning CTs from 23 patients was retrospectively analyzed. Anatomical deformation during treatment was estimated using free-form deformation (FFD) method from Elastix and DL-based VoxelMorph approaches. The VoxelMorph method was investigated using anatomical scans (VMorph_Sc) or label images (VMorph_Msk), or the combination of both (VMorph_Sc_Msk). Accumulated doses were compared with the planning dose. The DSC ranges, averaged for prostate, rectum and bladder, were 0.60-0.71, 0.67-0.79, 0.93-0.98, and 0.89-0.96 for the FFD, VMorph_Sc, VMorph_Msk, and VMorph_Sc_Msk methods, respectively. When including both anatomical and label images, VoxelMorph estimated more complex deformations resulting in heterogeneous determinant of Jacobian and higher percentage of deformation vector field (DVF) folding (up to a mean value of 1.90% in the prostate). Large differences were observed between DL-based methods regarding estimation of the accumulated dose, showing systematic overdosage and underdosage of the bladder and rectum, respectively. The difference between planned mean dose and accumulated mean dose with VMorph_Sc_Msk reached a median value of +6.3Gy for the bladder and -5.1Gy for the rectum. The estimation of the deformations using DL-based approach is feasible for male pelvic anatomy but requires the inclusion of anatomical contours to improve organ correspondence. High variability in the estimation of the accumulated dose depending on the deformable strategy suggests further investigation of DL-based techniques before clinical deployment.

Deep SE-BiLSTM with IFPOA Fine-Tuning for Human Activity Recognition Using Mobile and Wearable Sensors.

Pervasive computing, human-computer interaction, human behavior analysis, and human activity recognition (HAR) fields have grown significantly. Deep learning (DL)-based techniques have recently been effectively used to predict various human actions using time series data from wearable sensors and mobile devices. The management of time series data remains difficult for DL-based techniques, despite their excellent performance in activity detection. Time series data still has several problems, such as difficulties in heavily biased data and feature extraction. For HAR, an ensemble of Deep SqueezeNet (SE) and bidirectional long short-term memory (BiLSTM) with improved flower pollination optimization algorithm (IFPOA) is designed to construct a reliable classification model utilizing wearable sensor data in this research. The significant features are extracted automatically from the raw sensor data by multi-branch SE-BiLSTM. The model can learn both short-term dependencies and long-term features in sequential data due to SqueezeNet and BiLSTM. The different temporal local dependencies are captured effectively by the proposed model, enhancing the feature extraction process. The hyperparameters of the BiLSTM network are optimized by the IFPOA. The model performance is analyzed using three benchmark datasets: MHEALTH, KU-HAR, and PAMPA2. The proposed model has achieved 99.98%, 99.76%, and 99.54% accuracies on MHEALTH, KU-HAR, and PAMPA2 datasets, respectively. The proposed model performs better than other approaches from the obtained experimental results. The suggested model delivers competitive results compared to state-of-the-art techniques, according to experimental results on four publicly accessible datasets.

A DEBIASING VARIATIONAL AUTOENCODER FOR DEFORESTATION MAPPING

Abstract. Deep Learning (DL) algorithms provide numerous benefits in different applications, and they usually yield successful results in scenarios with enough labeled training data and similar class proportions. However, the labeling procedure is a cost and time-consuming task. Furthermore, numerous real-world classification problems present a high level of class imbalance, as the number of samples from the classes of interest differ significantly. In various cases, such conditions tend to promote the creation of biased systems, which negatively impact their performance. Designing unbiased systems has been an active research topic, and recently some DL-based techniques have demonstrated encouraging results in that regard. In this work, we introduce an extension of the Debiasing Variational Autoencoder (DB-VAE) for semantic segmentation. The approach is based on an end-to-end DL scheme and employs the learned latent variables to adjust the individual sampling probabilities of data points during the training process. For that purpose, we adapted the original DB-VAE architecture for dense labeling in the context of deforestation mapping. Experiments were carried out on a region of the Brazilian Amazon, using Sentinel-2 data and the deforestation map from the PRODES project. The reported results show that the proposed DB-VAE approach is able to learn and identify under-represented samples, and select them more frequently in the training batches, consequently delivering superior classification metrics.

Drug Design and Disease Diagnosis: The Potential of Deep Learning Models in Biology

Abstract: Early prediction and detection enable reduced transmission of human diseases and provide healthcare professionals ample time to make subsequent diagnoses and treatment strategies. This, in turn, aids in saving more lives and results in lower medical costs. Designing small chemical molecules to treat fatal disorders is also urgently needed to address the high death rate of these diseases worldwide. A recent analysis of published literature suggested that deep learning (DL) based models apply more potential algorithms to hybrid databases of chemical data. Considering the above, we first discussed the concept of DL architectures and their applications in drug development and diagnostics in this review. Although DL-based approaches have applications in several fields, in the following sections of the article, we focus on recent developments of DL-based techniques in biology, notably in structure prediction, cancer drug development, COVID infection diagnostics, and drug repurposing strategies. Each review section summarizes several cutting-edge, recently developed DL-based techniques. Additionally, we introduced the approaches presented in our group, whose prediction accuracy is relatively comparable with current computational models. We concluded the review by discussing the benefits and drawbacks of DL techniques and outlining the future paths for data collecting and developing efficient computational models.

Assessing the Impact of Image Resolution on Deep Learning for TB Lesion Segmentation on Frontal Chest X-rays.

Deep learning (DL) models are state-of-the-art in segmenting anatomical and disease regions of interest (ROIs) in medical images. Particularly, a large number of DL-based techniques have been reported using chest X-rays (CXRs). However, these models are reportedly trained on reduced image resolutions for reasons related to the lack of computational resources. Literature is sparse in discussing the optimal image resolution to train these models for segmenting the tuberculosis (TB)-consistent lesions in CXRs. In this study, we investigated the performance variations with an Inception-V3 UNet model using various image resolutions with/without lung ROI cropping and aspect ratio adjustments and identified the optimal image resolution through extensive empirical evaluations to improve TB-consistent lesion segmentation performance. We used the Shenzhen CXR dataset for the study, which includes 326 normal patients and 336 TB patients. We proposed a combinatorial approach consisting of storing model snapshots, optimizing segmentation threshold and test-time augmentation (TTA), and averaging the snapshot predictions, to further improve performance with the optimal resolution. Our experimental results demonstrate that higher image resolutions are not always necessary; however, identifying the optimal image resolution is critical to achieving superior performance.

Evaluation of Machine Learning based Network Attack Detection

The growth in the internet and communication technologies has driven tremendous developments in various application areas such as smart cities, cloud computing, internet-of-things, e-banking, e-commerce and e-government. However, with the advancements in networking infrastructure, hacking tools and methodologies have been much evolved thereby enabling hackers to attempt newer and more complicated cyber-attacks. Consequently, cyber-security has now emerged as a vital research area to address security concerns. Traditional security mechanisms such as firewalls and anti-viruses are not enough to protect networks and accurately detect intrusions. An Intrusion Detection System (IDS) provides an additional layer of security that prevents networks against possible intrusions through continuous surveillance of the network traffic. Despite the effectiveness of IDS and enormous research being conducted on the very topic, IDS still poses challenges to accurately detect intrusions, novel cyber-attacks and reducing false positive rates. Recently, Machine Learning (ML) and Deep Learning (DL) techniques have been exploited to overcome the inherent deficiencies of IDS. Existing research has demonstrated that ML and DL have great potential to detect intrusions and classify cyber-attacks in an efficient manner. Based on their inherent learning capabilities, ML and DL-based techniques can effectively detect patterns (features) from the network traffic and predict the behavior (normal or abnormal activity) based on these patterns. This research work first presents the concepts of IDS, followed by a comprehensive review of the recent ML and DL-based schemes. Later, a performance analysis of various ML algorithms is presented on a publicly available dataset to weigh their strengths and weaknesses in terms of accuracy and training time among others. We mainly evaluate the most commonly used supervised learning algorithms including Decision Trees (DT), Random Forest (RF), Gradient Booster (GB) and Neural Networks (NNs).

Deep learning: survey of environmental and camera impacts on internet of things images

Internet of Things (IoT) images are captivating growing attention because of their wide range of applications which requires visual analysis to drive automation. However, IoT images are predominantly captured from outdoor environments and thus are inherently impacted by the camera and environmental parameters which can adversely affect corresponding applications. Deep Learning (DL) has been widely adopted in the field of image processing and computer vision and can reduce the impact of these parameters on IoT images. Albeit, there are many DL-based techniques available in the current literature for analyzing and reducing the environmental and camera impacts on IoT images. However, to the best of our knowledge, no survey paper presents state-of-the-art DL-based approaches for this purpose. Motivated by this, for the first time, we present a Systematic Literature Review (SLR) of existing DL techniques available for analyzing and reducing environmental and camera lens impacts on IoT images. As part of this SLR, firstly, we reiterate and highlight the significance of IoT images in their respective applications. Secondly, we describe the DL techniques employed for assessing the environmental and camera lens distortion impacts on IoT images. Thirdly, we illustrate how DL can be effective in reducing the impact of environmental and camera lens distortion in IoT images. Finally, along with the critical reflection on the advantages and limitations of the techniques, we also present ways to address the research challenges of existing techniques and identify some further researches to advance the relevant research areas.