Research Challenges Research Articles

Interpretability Research of Deep Learning: A Literature Survey

Deep learning (DL) has been widely used in various fields. However, its black-box nature limits people's understanding and trust in its decision-making process. Therefore, it becomes crucial to research the DL interpretability, which can elucidate the model's decision-making processes and behaviors. This review provides an overview of the current status of interpretability research. First, the DL's typical models, principles, and applications are introduced. Then, the definition and significance of interpretability are clarified. Subsequently, some typical interpretability algorithms are introduced into four groups: active, passive, supplementary, and integrated explanations. After that, several evaluation indicators for interpretability are briefly described, and the relationship between interpretability and model performance is explored. Next, the specific applications of some interpretability methods/models in actual scenarios are introduced. Finally, the interpretability research challenges and future development directions are discussed.

Reversibly crosslinked polyurethane fibres from sugar‐based 5‐chloromethylfurfural: Synthesis, fibre‐spinning and fibre‐to‐fibre recycling

The development of recyclable crosslinked thermosetting fibres is a challenging research topic. In the present work, we have designed and synthesized polyurethane fibres from fructose‐derived 5‐chloromethylfurfural (CMF) and lignin‐derived monomeric phenols. The greenhouse gas emissions associated with the production of CMF showed comparable results to that of 5‐hydroxymethylfurfural (HMF), a high potential sugar‐based platform molecule. The wet‐spun biobased polyurethane fibres produced could be conveniently crosslinked using Diels–Alder chemistry to effectively enhance the glass transition temperature and mechanical properties. At a mildly elevated temperature (140 °C), the chemically crosslinked fibres could be effectively de‐crosslinked, which enabled complete separation from a mixture with poly(ethylene terephthalate) (PET) and cotton fibres. These results outline a potential strategy to design and fabricate new biobased fibres with reversible crosslinking, which may enable fibre‐to‐fibre recycling.

Different FeS Concentrations for Encapsulating ZIF-67 Nanomaterials toward the Enhanced Oxidation Evolution Reaction.

Due to the slow kinetic nature of the oxygen evolution reaction (OER), the development of electrocatalysts with high efficiency, stability, and economy for oxygen production using metal-organic framework (MOF) materials is still a challenging research topic. In this work, we chose the different concentrations of FeS adsorption to encapsulate metal cobalt-based ZIF-67 MOF for preparing a series of electrocatalysts (ZIF1FeSx, x = 0.2, 0.5, 0.75, and 1), which were mainly explored for the electrocatalytic OER. Among them, ZIF1FeS0.5 has excellent electrocatalytic activity for OER, which can be driven by low overpotentials of 276 and 349 mV at 10 and 50 mA cm-2 current densities, and more than 92% of the initial overpotential can be maintained after 100 h of continuous OER at 10 mA cm-2 current density. This is mainly due to the electronic interactions between the cobalt-based MOF and the FeS, which shift the electronic state of the active metal center to a higher valence state for increasing the number of active sites and enhancing the efficiency of electron transfer to facilitate the OER course. This work may contribute to the design of effective catalysts for the OER during the electrolysis of alkaline solutions.

Robust Face Recognition System Based on Deep Facial Feature Extraction and Machine Learning

Facial Recognition is a common challenge in research for security applications. Wide-ranging applications of this technology include biometrics, security data, access to controlled locations, smart cards, and surveillance systems. A practical approach to identifying individuals depends on factors such as partial face occlusion, illumination, age, expression, makeup, and poses. These complex face recognition variables affect most face recognition systems. In this paper, we propose a deep transfer learning for facial features based on the pose and illumination variations of celebrities’ faces in a free environment for face recognition. The new approach that combines deep learning methods for feature extraction with machine learning classification methods. Also, studies the performance of the pre-trained model (VGGFace2 with ResNet-50 model weights) for feature extraction with a Multilayer Perceptron Classifier (MLP), Decision Tree, and Bootstrap Aggregation (Bagging) to perform classification. The convolution neural network has lately achieved excellent advancement in facial recognition. The outcome indicated that VGGFace2 with MLP obtained more precision and provided the best results (Accuracy, F-measure, Recall, and Precision). The results for all metric for the proposed model is 99%. The benchmark for this model is a dataset of 105 people’s faces.

Anomaly detection and prediction of energy consumption for smart homes using machine learning

AbstractAs technology advances, smart homes are being increasingly adopted, thus generating massive data that pose new research challenges. We propose a machine learning framework for monitoring energy consumption in smart home devices. The proposed framework involves an anomaly detection module, followed by a predictive model to forecast energy consumption patterns in a typical smart home. We employ three outlier‐based techniques for anomaly detection: (1) local outlier factor, (2) connectivity‐based outlier factor, and (3) cluster‐based local outlier factor. Furthermore, we apply random forest, linear regression, decision tree, and the ensemble techniques of adaptive, gradient, and extreme gradient boosting to anomaly free data to develop baseline models that predict the energy consumption patterns of smart home devices. The framework is evaluated on three publicly available energy datasets collected from various smart homes. The experimental results reveal that the cluster‐based local outlier factor with extreme gradient boosting achieves promising results with high prediction accuracy.

Adaptation in Edge Computing: A Review on Design Principles and Research Challenges

Edge computing places the computational services and resources closer to the user proximity, to reduce latency, and ensure the quality of service and experience. Low latency, context awareness and mobility support are the major contributors to edge-enabled smart systems. Such systems require handling new situations and change on the fly and ensuring the quality of service while only having access to constrained computation and communication resources and operating in mobile, dynamic and ever-changing environments. Hence, adaptation and self-organisation are crucial for such systems to maintain their performance, and operability while accommodating new changes in their environment. This article reviews the current literature in the field of adaptive edge computing systems. We use a widely accepted taxonomy, which describes the important aspects of adaptive behaviour implementation in computing systems. This taxonomy discusses aspects such as adaptation reasons, the various levels an adaptation strategy can be implemented, the time of reaction to a change, categories of adaptation technique and control of the adaptive behaviour. In this article, we discuss how these aspects are addressed in the literature and identify the open research challenges and future direction in adaptive edge computing systems. The results of our analysis show that most of the identified approaches target adaptation at the application level, and only a few focus on middleware, communication infrastructure and context. Adaptations that are required to address the changes in the context, changes caused by users or in the system itself are also less explored. Furthermore, most of the literature has opted for reactive adaptation, although proactive adaptation is essential to maintain the edge computing systems’ performance and interoperability by anticipating the required adaptations on the fly. Additionally, most approaches apply a centralised adaptation control, which does not perfectly fit the mostly decentralised/distributed edge computing settings.

Toddlers and Adults Link Nouns to Typical Space, but What do Infants do?

ABSTRACT When adults hear a noun with an implicit up location, like “moon”, they typically respond faster with an up compared to a down response. Conversely, for nouns with an implicit down location, like “stone”, they respond faster with a down response. We test when respective language-space associations develop. German-learning infants (11–14 months), toddlers (2–4 years) and adults participated in an eye-tracking experiment. Participants first heard a noun implying an up or down location. We selected nouns that are likely to be known even by the infants. Following the spoken word, a target stimulus appeared either at the upper or lower screen part. In half of the trials the target appeared at the noun’s implicit position. In the other half of the trials the target appeared in the opposite position. Participants’ time to fixate the target (our preregistered measure) was not influenced by the previously heard nouns. We discuss methodological challenges of infant research that might have compromised the power needed for this preregistered analysis. However, our exploratory measure – participants’ fixations while listening to the implicit spatial words – revealed language-space association in adults and toddlers (but not in infants). Adults and toddlers fixated the upper part of the screen more than the lower part while listening to an “up word” and, conversely, the lower part more while listening to a “down word”. Thereby, our results provide evidence that children link nouns to typical space at least from the age of two.

Fine mapping of RNA isoform diversity using an innovative targeted long-read RNA sequencing protocol with novel dedicated bioinformatics pipeline.

Solving the structure of mRNA transcripts is a major challenge for both research and molecular diagnostic purposes. Current approaches based on short-read RNA sequencing and RT-PCR techniques cannot fully explore the complexity of transcript structure. The emergence of third-generation long-read sequencing addresses this problem by solving this sequence directly. However, genes with low expression levels are difficult to study with the whole transcriptome sequencing approach. To fix this technical limitation, we propose a novel method to capture transcripts of a gene panel using a targeted enrichment approach suitable for Pacific Biosciences and Oxford Nanopore Technologies platforms. We designed a set of probes to capture transcripts of a panel of genes involved in hereditary breast and ovarian cancer syndrome. We present SOSTAR (iSofOrmS annoTAtoR), a versatile pipeline to assemble, quantify and annotate isoforms from long read sequencing using a new tool specially designed for this application. The significant enrichment of transcripts by our capture protocol, together with the SOSTAR annotation, allowed the identification of 1,231 unique transcripts within the gene panel from the eight patients sequenced. The structure of these transcripts was annotated with a resolution of one base relative to a reference transcript. All major alternative splicing events of the BRCA1 and BRCA2 genes described in the literature were found. Complex splicing events such as pseudoexons were correctly annotated. SOSTAR enabled the identification of abnormal transcripts in the positive controls. In addition, a case of unexplained inheritance in a family with a history of breast and ovarian cancer was solved by identifying an SVA retrotransposon in intron 13 of the BRCA1 gene. We have validated a new protocol for the enrichment of transcripts of interest using probes adapted to the ONT and PacBio platforms. This protocol allows a complete description of the alternative structures of transcripts, the estimation of their expression and the identification of aberrant transcripts in a single experiment. This proof-of-concept opens new possibilities for RNA structure exploration in both research and molecular diagnostics.

Advancements in Energy Storage and Conversion: The Role of High‐Energy Irradiation Technology

AbstractContemporary challenges have emerged concerning the capability and safety of secondary batteries in the energy storage field. Additionally, catalysts employed in the energy conversion confront issues including inadequate stability and elevated costs. In addition to its traditional use, high‐energy irradiation has found extended application in controlled manipulation of materials for electrochemical energy storage and conversion, due to its unique advantages of no secondary pollution, convenience, and efficiency. This review article shows the recent progress of the application of high‐energy irradiation technology in the field of energy storage and conversion for the first time. The mechanism of material modification via high‐energy irradiation was elaborated, and the application of high‐energy irradiation in tailoring materials for battery components and catalysts was discussed in detail. Finally, challenges for future research of high‐energy irradiation technology in the field of energy storage and conversion were proposed. This review article will provide important significance for subsequent research on high‐energy irradiation technology in the field of energy storage and conversion.

Intelligent Prediction of Rate of Penetration Using Mechanism-Data Fusion and Transfer Learning

Rate of penetration (ROP) is crucial for evaluating drilling efficiency, with accurate prediction essential for enhancing performance and optimizing parameters. In practice, complex and variable downhole environments pose significant challenges for mechanistic ROP equations, resulting in prediction difficulties and low accuracy. Recently, data-driven machine learning models have been widely applied to ROP prediction. However, these models often lack mechanistic constraints, limiting their performance to specific conditions and reducing their real-world applicability. Additionally, geological variability across wells further hinders the transferability of conventional intelligent models. Thus, combining mechanistic knowledge with intelligent models and enhancing model stability and transferability are key challenges in ROP prediction research. To address these challenges, this paper proposes a Mechanism-Data Fusion and Transfer Learning method to construct an intelligent prediction model for ROP, achieving accurate ROP predictions. A multilayer perceptron (MLP) was selected as the base model, and training was performed using data from neighboring wells and partial data from the target well. The Two-stage TrAdaBoost.R2 algorithm was employed to enhance model transferability. Additionally, drilling mechanistic knowledge was incorporated into the model’s loss function as a constraint to achieve a fusion of mechanistic knowledge and data-driven approaches. Using MAPE as the measure of accuracy, compared with conventional intelligent models, the proposed ROP prediction model improved prediction accuracy on the target well by 64.51%. The model transfer method proposed in this paper has a field test accuracy of 89.71% in an oilfield in China. These results demonstrate the effectiveness and feasibility of the proposed transfer learning method and mechanistic–data integration approach.

Challenges in radioecology following the new trends in UAE's agriculture and environmental changes: a review.

The development of the nuclear industry in the countries of the Arabian Gulf demands an investigation of its potential impact on the environment and human activities. It should involve routine monitoring of radionuclide from existing nuclear facilities as well as modelling of accidental release of radioactivity. Agriculture is usually considered the human activity which is the most endangered with radioactive pollution. Although the traditional concept of low-developed agriculture in the United Arab Emirates has been already described from a radioecological point of view, herein, the recent trends of agriculture are recorded. Due to climatic changes and population increase, the use of saline lands will be more pressing. Apart from it, the risk assessment for nuclear events, which was formed mainly for temperate zones, neglected another aspect of arid zones which is underground water reservoirs refilled with episodic heavy rains. The sandy or saline soils are not an effective barrier for the migration of radionuclide deposited in topsoil. A hypersaline environment could prevent the adsorption of radionuclide on soil particles and allow their high bioavailability for halophyte plants. The new principal challenges in radioecological research in arid zones are impacts on (i) saline agriculture, (ii) underground water reservoirs, and (iii) proposal for countermeasures in order to minimise the impact of radioactive contamination.

Computationally efficient deep learning models for diabetic retinopathy detection: a systematic literature review

Diabetic retinopathy, often resulting from conditions like diabetes and hypertension, is a leading cause of blindness globally. With diabetes affecting millions worldwide and anticipated to rise significantly, early detection becomes paramount. The survey scrutinizes existing literature, revealing a noticeable absence of consideration for computational complexity aspects in deep learning models. Notably, most researchers concentrate on employing deep learning models, and there is a lack of comprehensive surveys on the role of vision transformers in enhancing the efficiency of these models for DR detection. This study stands out by presenting a systematic review, exclusively considering 84 papers published in reputable academic journals to ensure a focus on mature research. The distinctive feature of this Systematic Literature Review (SLR) lies in its thorough investigation of computationally efficient approaches and models for DR detection. It sheds light on the incorporation of vision transformers into deep learning models, highlighting their significant contribution to improving accuracy. Moreover, the research outlines clear objectives related to the identified problem, giving rise to specific research questions. Following an assessment of relevant literature, data is extracted from digital archives. Additionally, in light of the results obtained from this SLR, a taxonomy for the detection of diabetic retinopathy has been presented. The study also highlights key research challenges and proposes potential avenues for further investigation in the field of detecting diabetic retinopathy.

A Novel Parameter-Variabled and Coupled Chaotic System and Its Application in Image Encryption with Plaintext-Related Key Concealment

The design of a chaotic system and pseudo-random sequence generation method with excellent performance and its application in image encryption have always been attractive and challenging research fields. In this paper, a new model of parameter-variabled coupled chaotic system (PVCCS) is established by interaction coupling between parameters and states of multiple low-dimensional chaotic systems, and a new way to construct more complex hyperchaotic systems from simple low-dimensional systems is obtained. At the same time, based on this model and dynamical DNA codings and operations, a new pseudo-random sequence generation method (PSGM-3DPVCCS/DNA) is proposed, and it is verified that the generated pseudo-random sequence of PSGM-3DPVCCS/DNA has excellent random characteristics. Furthermore, this paper designs a novel pixel chain diffusion image encryption algorithm based on the proposed parameter-variabled coupled chaotic system (PVCCS) in which the hash value of plaintext image is associated with the initial key to participate in the encryption process so that the encryption key is closely associated with plaintext, which improves the security of the algorithm and effectively resists the differential cryptanalysis risk. In addition, an information hiding method is designed to hide the hash value of plaintext image in ciphertext image so that the hash value does not need to be transmitted in each encryption, and the initial key can be reused, which solves the key management problem in application and improves the application efficiency of the encryption algorithm. The experimental analysis shows that the chaotic system constructed in this paper is creative and universal and has more excellent chaotic characteristics than the original low-dimensional system. The sequence generated by the pseudo-random sequence generation method has excellent pseudo-random characteristics and security, and the image encryption algorithm can effectively resist differential cryptanalysis risk, showing advanced encryption performance.

External Thermal Insulation Composite Systems—Past and Future in a Sustainable Urban Environment

In recent decades, the sustainable development of the planet has been negatively affected by a number of factors, including the construction industry. The construction industry includes, among other things, the highly topical energy reconstruction of existing prefabricated residential housing, which is implemented to improve their condition from a thermal engineering and energy perspective. Composite materials, known as external thermal insulation composite systems (ETICSs), have come to the fore, bringing a number of undeniable benefits to society. After more than 20 years of experience, it turns out that in addition to the benefits, ETICSs also bring new research challenges to the discussion, which are related to the issue of the biocorrosion of the external envelope of ETICSs, and also to the issue of the indoor microclimate. Based on the literature review and case studies, we aim to show that ecologically friendly building materials require a multidisciplinary approach. At the same time, we want to contribute to the discussion of whether the diversity of microorganisms on ETICS composites is a potential source of health risks and whether the transport of microorganisms to the indoor environment can be ruled out through natural ventilation from the outdoor environment to the interior.

Progress in Research on Terpenoid Biosynthesis and Terpene Synthases of Lauraceae Species

Lauraceae, an important family of Angiospermae, comprises over 2500 species widely distributed in tropical and subtropical evergreen broad-leaved forests. This family is renowned for its rich resource of terpenoids, particularly monoterpenes, sesquiterpenes, and diterpenes. These compounds not only impart specific scents to Lauraceae species but also play crucial roles in plant growth, development, and environmental adaptation. These compounds also possess extensive bioactivities, such as antioxidant, antibacterial, anti-inflammatory, and neuroprotective effects, making them valuable in the fields of perfumery, cosmetics, food, and medicine, and thus holding significant economic value. Recent advancements in high-throughput technologies, especially genomics, transcriptomics, and metabolomics, have significantly advanced our knowledge of the chemical constituents and biosynthetic pathways of terpenoids in Lauraceae species. Such progress has also shed light on the diversity and functionality of the terpene synthases (TPSs) gene family, a key enzyme involved in terpenoid biosynthesis. This paper reviews the latest research findings on the biosynthetic pathways of terpenoids and their key enzyme-encoding gene families in Lauraceae plants. We also analyze the evolutionary patterns of TPS gene family members of four Lauraceae species at the whole-genome level and summarize their mechanisms of action in secondary metabolite synthesis. Furthermore, this paper highlights the current research challenges and proposes prospects, such as the complexity of gene families, the uncertainties in functional predictions, and unclear regulatory mechanisms. Our objective is to provide scientific foundations for the in-depth analysis of terpenoid biosynthesis mechanisms and the development and utilization of natural products in Lauraceae plants.

Full-Azimuth Quantitative Identification of Near-Fault Pulse-Like Ground Motions Based on M&P Wavelet

ABSTRACT Quantitative characterization and proper parameterization of pulse-like ground motions have been a challenge for earthquake engineering research. The azimuth of the pulse-like ground motion has a great influence on the structural damage. This study proposes a full-azimuth quantitative identification method to determine the pulse inherent in near-fault pulse-like ground motions. A support vector machine was used to classify the records into pulses or non-pulses based on a new pulse index. Records with a pulse index greater than zero can be classified as pulse-like. The proposed approach was used to identify pulse-like features in arbitrary orientations and can determine the strongest pulse azimuth.

Addressing Constraint Coupling and Autonomous Decision-Making Challenges: An Analysis of Large-Scale UAV Trajectory-Planning Techniques

With the increase in UAV scale and mission diversity, trajectory planning systems faces more and more complex constraints, which are often conflicting and strongly coupled, placing higher demands on the real-time and response capabilities of the system. At the same time, conflicts and strong coupling pose challenges the autonomous decision-making capability of the system, affecting the accuracy and efficiency of the planning system in complex environments. However, recent research advances addressing these issues have not been fully summarized. An in-depth exploration of constraint handling techniques and autonomous decision-making issues will be of great significance to the development of large-scale UAV systems. Therefore, this paper aims to provide a comprehensive overview of this topic. Firstly, the functions and application scenarios of large-scale UAV trajectory planning are introduced and classified in detail according to the planning method, realization function and the presence or absence of constraints. Then, the constraint handling techniques are described in detail, focusing on the priority ranking of constraints and the principles of their fusion and transformation methods. Then, the importance of autonomous decision-making in large-scale UAV trajectory planning is described in depth, and related dynamic adjustment algorithms are introduced. Finally, the future research directions and challenges of large-scale UAV trajectory planning are outlooked, providing directions and references for future research in the fields of UAV clustering and UAV cooperative flight.

Exploring Rheumatoid Arthritis: Unraveling Signaling Pathways, Advancing Management, and Charting Future Perspectives

Rheumatoid arthritis (RA) is a chronic autoimmune disorder in which inflammation occurs in the synovial joint and synovial membrane, and in severe condition, it leads to erosion of cartilage and bone. The etiology of RA is still indescribable; generally, RA is a multifactorial disease, wherein complex interactions between patients and different factors such as environmental, genetic, and epigenetic factors determine the overall risk of disease. RA is classified based on the stage of disease. Treatment for RA is still a challenge for doctors and researchers. Effective treatment for RA can be achieved by decreasing disease activity or reduction of RA to other patients using currently available treatment options. This article gives a comprehensive perspective of data regarding different signaling pathways involved in RA development, factors affecting RA, stages of RA, and other treatments of RA. This article also suggests future research perspectives that would be helpful for the development of RA treatment (medicine) with more efficacy, safety profiles, and lower financial costs.

Non-coding RNA networks in pulmonary arterial hypertension.

Pulmonary artery hypertension (PAH) is a severe cardiovascular disease marked by a persistent increase in pulmonary artery resistance and pressure, leading to right ventricular strain, hypertrophy, and eventually right heart failure and death. Despite numerous available targeted therapies, the clinical needs for treating PAH remain unmet. Current treatments primarily aim to dilate pulmonary vessels rather than reverse pulmonary vascular remodeling, failing to offer a fundamental solution for PAH. Therefore, developing new therapies for this condition is urgently required. Recent research has highlighted the crucial role of non-coding RNAs (ncRNAs) in the occurrence and development of PAH. ncRNAs, such as long non-coding RNAs(lncRNA), circular RNAs (circRNA), microRNAs (miRNA), and PIWI-interacting RNAs (piRNA), are a class of transcripts that do not translate proteins but affect various diseases at different levels, including chromatin modification, transcription regulation, post-translational processes. The current study delves into recent advancements in understanding how lncRNAs, circRNAs, miRNAs, and piRNAs contribute to the pathogenesis of PAH. This review addresses the existing research challenges and explores the potential of ncRNAs as both biomarkers and therapeutic targets, suggesting that ncRNAs may serve as valuable indicators and treatment options for the disease.

Predictive process monitoring: concepts, challenges, and future research directions

Predictive Process Monitoring (PPM) extends classical process mining techniques by providing predictive models that can be applied at runtime during the execution of a business process, for example, to predict the sequence of the next event(s) in a case, its outcomes, or performance-related aspects such as the remaining processing time. These predictive models go beyond process mining’s inherent descriptive nature that is offered by typical process discovery, conformance checking, and model enhancement techniques. The growing interest in PPM is driven by its additional value proposition, i.e., delivering real-time information regarding the future execution of business process instances, thus allowing for better informed operational decision making and performance analysis. The rapid growth of PPM during the last ten years has left the field lacking a cohesive taxonomy and an explicit recognition of the prevailing challenges. This paper aims at closing this gap, by comprehensively defining PPM using a unified approach to the key terms and by discussing challenges and opportunities in the field. Specifically, we propose three overarching research challenges and nine research directions, which have been validated through a survey with PPM researchers.