The following International Journal of Medical Engineering and Informatics article, "Endoscopic ultrasonography in differential diagnostics of benign and malignant pathology of the common bile ducts utilising fuzzy mathematical technologies", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
A study of advertising embedded in Indian reality television examines how stealth, or subliminal, advertising, affects brand recall in a media environment saturated with promotional messages. The work, published in the International Journal of Indian Culture and Business Management, looks at how audiences exposed to thousands of brand impressions daily are affected by this approach to advertising. The findings have implications for marketing departments, which are increasingly using subtle product placement to familiarise potential and known customers with brands.
The research used within-subject experiments where participants viewed 12-minute clips from five Indian reality programmes containing embedded brand appearances. Recall was measured using standardised questionnaires assessing brand memory after exposure. Across 20 brands, most achieved moderate recall through this form of advertising.
The team found that two factors most influenced brand recall: whether the placement was spoken or visual and whether it meshed with the premise or arc of the segment. The results showed that verbal references were more effective than visual placements, but higher congruity between brand and the storyline of the segment improved recall.
Perhaps paradoxically, visual prominence of a product did not affect brand recall significantly. This challenges the marketing assumption that visibility drives effectiveness. The researchers concluded that embedded advertising depends largely on more subtle cues than straightforward visibility in the segment. Product placement is on the rise across TV, streaming platforms, and in digital content. Marketing departments hoping to familiarise consumers with their products need to understand these findings to benefit the most.
Dua, G.G., Vaidya, R.D. and Bapat, G.S. (2026) ‘Unveiling the recall power of covert advertising: a study on stealth advertising in Indian reality shows’, Int. J. Indian Culture and Business Management, Vol. 38, No. 1, pp.68–93.
The following International Journal of Business Innovation and Research article, "Fuelling the customer experience: an exploration of service quality in fuel stations through importance-performance analysis", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
Researchers have developed an artificial intelligence model that predicts crime more accurately than several existing approaches by combining information about where crimes occur, when they happen, and wider social patterns. They report details of the approach in the International Journal of Innovative Computing and Applications.
The model combines a graph convolutional network, which identifies relationships between locations, with a transformer, an AI architecture designed to detect patterns over time. Together, the techniques allow the system to capture both spatial and temporal trends in criminal activity. The researchers also incorporated a generative adversarial network (GAN), a system in which two AI models compete to improve performance. The GAN was enhanced using a variational autoencoder, a method that helps generate more representative data while reducing common training problems such as biased outputs and vanishing gradients, where learning slows or stops.
The system integrates several machine-learning techniques to analyse complex datasets that traditional methods often struggle to process. In tests on historical data from several US cities, including Los Angeles and Seattle, the model achieved an accuracy rate of 86.3 per cent when predicting robberies. The strongest competing systems had an accuracy of 83.2 per cent. The new system also gave strong results across other crime categories.
The researchers suggest that accurate forecasting could help law enforcement allocate resources more effectively and identify areas at higher risk of crime. However, there are limitations. The system was less accurate in areas with sparse crime data and struggled to make predictions in locations with little or no historical information. Future work will focus on adapting the model to such environments through transfer learning, which will allow knowledge gained in one setting to be applied to another.
Xie, M. (2026) ‘Multidimensional crime prediction technique optimisation combining feature extraction and GAN’, Int. J. Innovative Computing and Applications, Vol. 15, No. 5, pp.289–299.
The following Progress in Computational Fluid Dynamics, An International Journal article, "Quantifying wall effects on spherical particle settling with the Lattice Boltzmann method", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
The following International Journal of Continuing Engineering Education and Life-Long Learning article, "Classroom student emotion recognition using an improved segmentation clustering and multi-feature fusion emotion recognition algorithm", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
The following International Journal of Power and Energy Conversion article, "Change detection framework for power facilities in disaster scenarios", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
Researchers have developed an artificial intelligence-based patient monitoring system they say can identify signs of clinical deterioration earlier and more accurately than existing approaches. The system could help hospital staff intervene before a patient’s condition becomes critical. Details are discussed in the International Journal of Ad Hoc and Ubiquitous Computing.
Traditional monitoring systems rely largely on fixed thresholds for individual measurements such as heart rate, blood pressure or oxygen levels. However, these approaches often fail to account for differences between patients and may overlook how physiological changes interact across the body.
The new approach combines three machine-learning techniques. An adaptive attention mechanism continuously adjusts the importance assigned to different physiological signals. A spatiotemporal graph neural network analyses how vital signs influence one another and evolve. The system also incorporates reinforcement learning, a method in which algorithms learn decision-making strategies through feedback, enabling it to provide active clinical decision support rather than simply issuing alarms.
Tests were carried out to see how well the system performed in predicting historical outcomes recorded in two major intensive care unit (ICU) databases, MIMIC-III and eICU. The system achieved 96.3 per cent anomaly detection accuracy, generated warnings almost 40 minutes before critical events occurred, and reduced false alarms to 6.4 per cent.
Cheng, S., Zhu, J., Guan, S., Cheng, J. and Dou, T. (2026) ‘Intelligent monitoring of patient vital signs based on adaptive attention fusion spatiotemporal graph neural network’, Int. J. Ad Hoc and Ubiquitous Computing, Vol. 52, No. 5, pp.48–62.
The following International Journal of Systems, Control and Communications article, "Automation measures and power quality optimisation of power line carrier control based on electric spring and wavelet transform", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
The following International Journal of Technology Enhanced Learning article, "Unlocking computational thinking: immersive technologies for solving complex problems", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
A machine-learning model based on Transformer architecture, a form of artificial intelligence originally developed for language processing, can be used to detect heart disease from electrocardiogram (ECG), according to research in the International Journal of Medical Engineering and Informatics. Tests show it works well with data from several well-known medical datasets.
Heart disease is a major healthcare problem, with almost 18 million dying prematurely each year because of it. The challenge if finding ways to detect cardiovascular disease early enough to make treatment effective. An ECG is the standard way to record the heart’s electrical activity and is thus a common diagnostic tool. Interpretation of the trace needs significant expertise, is time-consuming and is not without the risk of misinterpretation.
The researchers discusses a one-dimensional (1D) Transformer model that can analyse ECG signals and in parallel with other clinical data. In tests, it was up to 94.2 per cent accurate in spotting the early stages of heart disease. Such precision coupled with expert clinical assessment can suffice to give the healthcare team more reliable options in taking a patient on to the next step in diagnosis and potential treatment.
The researchers suggest that their approach needs further development and validation with independent clinical datasets before it can be tested in a live clinical setting.
Miloud Aouidate, A. (2026) ‘Heart disease detection using 1D transformer network: case of ECG signals and clinical data’, Int. J. Medical Engineering and Informatics, Vol. 18, No. 5, pp.1–18.
A comparative modelling study of two Brazilian rain catchments suggests that climate change will have contrasting effects on future soil erosion in the Amazon and Cerrado. The findings have implications for land management in both biomes in the coming decades.
The research, published in the International Journal of Hydrology Science and Technology, used rainfall records and projections from NASA’s Earth Exchange Global Daily Downscaled Projections (NEX-GDDP). This allowed the team to estimate soil loss under different scenarios.
Soil loss, the amount of soil removed by erosion, was calculated using the Universal Soil Loss Equation (USLE), which estimates erosion driven by rainfall, land cover, topography, and soil properties. Rainfall erosivity refers to rain’s capacity to detach and transport soil particles.
The team explains that the Cerrado is a biodiverse savanna, while the Amazon regulates rainfall through evapotranspiration, the transfer of water from land to the atmosphere. Differences in rainfall, elevation and land use affect soil erosion patterns. The global circulation models (GCMs) of climate simulations suggest that shifting rainfall intensity will change erosion rates with heavier rainfall increasing soil loss risk.
In the Amazon catchment, soil loss historically ranges from just a few dozen kilograms per hectare each year to almost 20 tonnes per hectare per year. Soil erosion is predicted to increase by several per cent through the remainder of this century.
By contrast, soil loss in the Cerrado’s Piranhas River catchment has been from a hundred kilograms or so to almost 250 tonnes per hectare per year. The models predict that soil erosion will actually become less of a problem in the coming years despite the effects of climate change.
As such, there is a need to respond to the changes of the future in different ways in the two very different regions. In the Amazon catchment, rising erosion risk supports conservation measures such as terracing, no-till farming and crop rotation. While in the Cerrado catchment, simply maintaining current agricultural practices may help sustain reduced soil loss.
Sobral, R.V.S., Lobato, A.K.R., Soares, A.d.C.L., de Mendonça, L.M., Cruz, J.d.S. and Blanco, C. (2026) ‘Impact of climate change on soil loss in small catchments in the Amazon and Cerrado biomes’, Int. J. Hydrology Science and Technology, Vol. 21, No. 4, pp.417–436.
A new AI system can convert social media discussion about a product into a new design that takes into account user needs more accurately than earlier approaches, according to research in the International Journal of Information and Communication Technology.
The work addresses the problem of complexity in attempting to extract useful information from social network data for product development. Comments and reviews are typically unstructured, meaning they do not follow a fixed format, and also have many variables, such as sentiment, context, and usage scenarios, which makes it difficult to translate into insights about how people feel about products.
A deep-learning framework is at the heart of the system and combines various AI components. Firstly, it uses a multi-scale attention network to identify emotional needs in user comments. Attention in machine learning refers to a mechanism that prioritises the most relevant information in a dataset. The idea of multi-scale processing means it captures both detailed and broad patterns in language. The second component is a generative adversarial network (GAN). This uses two models working against each other, with one generating images and the other evaluating them. In addition, a spatial cross-reconstruction module refines image features, while a semantic correlation module links textual emotion signals to visual attributes. All of this works to improve the link between what the users say about the original product and the new design.
In tests, the model achieved more than 90 per cent accuracy in identifying the users’ emotional needs. This improves on existing methods and suggests that AI might help with data-driven product design informed by user sentiment and social media behaviour.
Wang, C. (2026) ‘Deep learning-based innovative product design driven by social network data’, Int. J. Information and Communication Technology, Vol. 27, No. 49, pp.59–78.
The following World Review of Entrepreneurship, Management and Sustainable Development article, "Assessing environmentally sustainable practices in boutique hotels: frontline staff perspectives", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
A study in the Global Business and Economics Review suggests that the failure of small and medium-sized enterprises (SMEs) can be predicted as much as three years before insolvency. The work could offer lenders, investors, and business owners an early warning of financial problems years in advance.
The researchers analysed data from more than 24500 European companies over an eight-year period. From this data, they developed a forecasting model that has an overall accuracy of about 82 per cent. It could identify more than 70 per cent of insolvencies three years in advance on test data with known outcomes. The final model relies on seven financial indicators: cash ratio, contribution per interest paid ratio, solvency ratio, short-term financing, leverage, debt-assets ratio, and return on assets. These measures capture a company’s liquidity, debt burden, financial resilience, and profitability. However, the model could yet be improved if there were greater disclosure from SMEs. That said, this is highly unlikely given the nature of smaller businesses.
The researchers say the work addresses a big gap in the corporate finance literature. Traditionally, this has focused on large publicly listed companies. However, SMEs account for most businesses in OECD economies and roughly two-thirds of employment, making their stability an important economic issue.
Silva, S. (2026) ‘Corporate failure prediction model for European SMEs’, Global Business and Economics Review, Vol. 34, No. 4, pp.395–419.
The following International Journal of Electronic Security and Digital Forensics article, "Judgement stage in electronic administrative proceedings and evidentiary authority", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
The following International Journal of Entrepreneurship and Small Business article, "Roots of innovative knowledge in small commercial enterprises in Ecuador", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
Research in the International Journal of Information and Communication Technology has taken inspiration from the hunting behaviour of the fish-eating bird of prey, the osprey, and combined this with inspiration from quantum computing to improve machine translation, particularly for long sentences and technical texts between Chinese and English.
Ospreys scan large areas of the water before making precise dives on their piscine targets. This strategy has been modelled and adapted into an algorithm that balances broad exploration with focused searches for promising solutions. The result in this work is the Quantum-Optimised Osprey Optimisation Algorithm (QOOA). The team explains that QOOA uses qubits, the mathematical units of quantum information, to explore a wider range of possible solutions. It also incorporates a quantum rotation mechanism that shifts from broad exploration to targeted refinement as the search progresses.
The team tested the new model on the WMT2018 English-Chinese translation benchmark, which contains almost 177,000 training examples. Compared with a baseline neural machine translation system, QOOA scored 3.2 percentage points higher and reduced the number of post-translation edits needed by 12.7 per cent. In addition, the team reports that their approach was particularly effective for lengthy and technical texts, where previous translation systems have been prone to errors and ambiguity.
Wang, L. (2026) ‘Application of quantum optimisation osprey algorithm in English translation quality improvement model’, Int. J. Information and Communication Technology, Vol. 27, No. 49, pp.1–18.
Microbial fuel cells (MFCs), which use microorganisms to generate electricity from organic waste, are emerging as a tool in the transition to cleaner energy systems and for the treatment of waste water, according to a review of recent research in the International Journal of Environment and Waste Management.
Unlike conventional power generation, MFCs use bacteria that break down organic matter and generate electrons as part of their natural metabolism. These electrons can be tapped off from the fuel cell by electrodes to create an electrical current. The review points out that wastewater, food waste, and agricultural by-products can all be used as a food supply for the bacteria and therefore as a sustainable fuel source for power production.
Indeed, the researchers argue that the greatest strength of this technology is to combine electricity generation with waste treatment. In wastewater facilities, MFCs can help remove organic pollutants while simultaneously producing power, potentially reducing the energy demands of treatment plants.
The team highlights advances in electrode materials, including carbon nanotubes, graphene, and conductive polymers. The review also considers the role of electroactive bacteria. These are microbes that can transfer electrons directly to the electrodes and include those in the Geobacter, Shewanella, and Pseudomonas genera.
Challenges remain, however. Power output is still relatively low in MFCs, and scaling systems from the laboratory bench to an industrial operation remains difficult. Cost, efficiency and long-term reliability must improve considerably to allow MFCs to achieve widespread commercial adoption.
Deshmukh, S.M., Dhokpande, S.R. and Sankhe, A.A. (2026) ‘Innovative applications and recent developments in microbial fuel cells: a comprehensive review’, Int. J. Environment and Waste Management, Vol. 40, No. 1, pp.1-26.
The following International Journal of Environmental Engineering article, "Low-carbon interior decoration lifecycle analysis based on BIM technology", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
The following International Journal of Medical Engineering and Informatics article, "Heart disease detection using 1D transformer network: case of ECG signals and clinical data", is freely available for download as an open access article.
It can be downloaded via the full-text link available here.
Research in the International Journal of Information and Communication Technology has looked at potential security and privacy weaknesses in remote higher education systems, focusing on centralised virtual learning platforms.
The researchers explain that these platforms usually rely on a single administrative infrastructure for authentication, records, and content delivery. This, they suggest, creates a single point of failure, where disruption or compromise of the central system might then affect the entire environment. This could open up the possibility of data tampering, credential fraud, and unauthorised access, while undermining trust in online degrees.
The team suggests that blockchain technology, usually associated with digital, or crypto, currencies, has the potential to protect education systems, making them tamper-proof. Earlier work has been tried allowing simple static credential storage. But the new approach is dynamic and could be used for enrolment, course access and ongoing assessment, rather than being mainly a certificate verification system.
The team’s proposal of Blockchain-Enabled Secure Distance Learning (BESDL) represents a lifecycle-based framework covering the full education process. It uses smart contracts, self-executing rules on a blockchain, to manage decentralised identity management, secure content-based access control, and encrypted content delivery.
Tests suggest improved authentication speed, better security, and greater scalability under high concurrent student loads compared with conventional systems.
Chen, J. and Chang, X. (2026) ‘Blockchain-enabled secure distance learning platforms for higher education‘, Int. J. Information and Communication Technology, Vol. 27, No. 56, pp.1-31.
A waste-to-energy system designed for palm oil mills could turn agricultural waste into electricity, industrial fuels and carbon-storing materials while generating commercially viable returns, according to an “Industrial Note” in the International Journal of Agriculture Innovation, Technology and Globalisation.
The authors examined Factory in Factory (FiF) systems wherein an integrated biomass treatment system built around two linked technologies can be used to convert organic waste into usable energy and saleable by-products. The work argues that the approach could help industries reduce greenhouse gas emissions while addressing mounting pressure on landfill capacity and waste disposal.
The system is aimed particularly at palm kernel cake (PKC). This is a waste residue from palm oil production that is generated in vast quantities at mills across Malaysia. The material is already concentrated at these industrial sites, so using FiF means transportation and collection costs are avoided almost entirely.
At the centre of the process is Pyrolysis Molecularisation Extraction Technology (PMET), which uses pyrolysis. Pyrolysis is the thermal decomposition of organic material without oxygen. This approach can process around 300 kilograms of biomass per hour. The process generates combustible gas, carbon-rich biochar and a liquid bio-oil known as green tar.
Biochar, a charcoal-like substance, can be used either to sequester carbon for long periods or as a soil improver and for pollution treatment. The bio-oil could be used as industrial fuel or as a feedstock for chemical, pharmaceutical, and biomedical products as an alternative to fossil products from the petrochemical industry.
The authors explain that a second component, the Gas Generator Assemble Cabinet (GGAC), can use pyrolysis-generated gas in electricity production. Such units can generate around 130 megawatts of electricity per month. This would allow mills either to offset their own power use or sell electricity to the national grid.
Lee, C-W. and Kao, W. M-W. (2026) ‘Industry note: Innovative waste-to-energy pyrolysis technology for sustainable biowaste utilisation‘, Int. J. Agriculture Innovation, Technology and Globalisation, Vol. 5, No. 2.
Research in the International Journal of Information and Communication Technology discusses the development of an artificial intelligence (AI) system that combines text, images and reviewer behaviour to detect and trace fake e-commerce reviews. The system could address the growing challenge faced by online marketplaces as deceptive feedback becomes increasingly sophisticated.
The team used a multimodal approach to analyse several types of data at once rather than relying solely on an examination of written comments. Existing systems often focus on review text or simple behavioural indicators, making them vulnerable to fabricated reviews paired with misleading images.
To improve detection, the researchers used a text convolutional neural network. This is a machine-learning model designed to identify patterns in language. In parallel, a pre-trained language model was employed that captures broader semantic meaning. The team adds that information about reviewers was also incorporated into the analysis as well as images attached to reviews. The images were analysed using a residual network, a deep-learning architecture used in computer vision.
The system then brings together these various signals to work out whether a particular review is genuine or not. A Transformer model, widely used in modern AI systems, could then be used to trace the origins and spread of a review flagged as suspicious. Tests on large-scale datasets showed measurable gains over existing methods, the team reports.
Duan, B. (2026) ‘Precise identification and traceability of fake e-commerce reviews integrating multimodal semantic understanding’, Int. J. Information and Communication Technology, Vol. 27, No. 35, pp.81–102.
Researchers have developed a new artificial intelligence-based system designed to improve cyberattack detection in software-defined networks (SDNs), a networking architecture widely used in data centres and enterprise systems.
The system combines a deep quantum neural network with a novel optimisation technique inspired by the behaviour of wild geese and dwarf mongooses. Its aim is to identify abnormal network traffic, including distributed denial-of-service (dDoS) attacks, while preventing network controllers from becoming overloaded.
SDNs differ from traditional networks by separating the control plane, which makes routing decisions, from the data plane, which forwards traffic. While this design improves flexibility and centralises management, it also creates potential targets for attackers seeking to disrupt communications between controllers and network devices.
In the new approach outlined in the International Journal of Heavy Vehicle Systems, network traffic is analysed using a deep quantum neural network, a machine-learning model designed to recognise complex patterns. When suspicious traffic is detected, the system assesses controller workloads and automatically transfers network switches from overloaded controllers to those with spare capacity.
In simulations, the researchers demonstrated a detection accuracy of 93.7%. They obtained a true positive rate of 91.6% and a true negative rate of 87.5%. The researchers argue that combining traffic anomaly detection with automated load balancing could strengthen increasingly centralised network infrastructures.
Ahsan Shariff, M. and Nelson Kennedy Babu, C. (2026) ‘Traffic anomaly detection with wild geese dwarf mongoose optimisation_DQNN’, Int. J. Heavy Vehicle Systems, Vol. 33, No. 2, pp.147–172.
Researchers have developed an enhanced wearable motion-tracking system that could improve the accuracy of fitness trackers used to monitor exercise and training. The team provides details in the International Journal of Data Mining and Bioinformatics.
Current wearable devices often show inconsistencies in heart-rate monitoring and can miscalculate calories burnt, speed, and distance travelled. Such inaccuracies limit their usefulness for health-conscious consumers, but particularly for athletes and their coaches who need precision.
The new work hopes to improve both data collection and sensor calibration. Researchers used fuzzy algorithms, computational methods designed to handle uncertain or variable information, to analyse real-time exercise data. They also applied filtering techniques to remove noise and improve data quality before calibrating the device’s sensors.
In their tests, they found that measurements of heart rate, calorie expenditure, movement speed, and distance closely matched those obtained through standard laboratory procedures. The researchers suggest that their main advance lies in combining improved sensor calibration with more sophisticated data processing. This allows the device to generate a more reliable picture of an athlete’s training performance in real time.
The findings could be used beyond competitive sport to help users develop personalised fitness programmes for health monitoring and injury prevention by giving them more dependable information about their physical activity.
Wu, F., Yang, S., Zhang, C. and Wu, H. (2026) ‘Application of wearable motion tracking devices in training, monitoring, and evaluation’, Int. J. Data Mining and Bioinformatics, Vol. 30, No. 6, pp.71–91.
An artificial intelligence (AI) system that combines breast cancer tissue images with molecular marker data achieves high accuracy in diagnosis, tumour classification, and survival prediction. Details are reported in the International Journal of Data Mining and Bioinformatics.
A common limitation of breast cancer care is that medical imaging and molecular markers as well as hormone receptor status are usually analysed separately. The researchers suggest that this can reduce the effectiveness of early detection, subtype classification, and personalised treatment planning. Their new addresses this issue.
In testing, the system achieved an accuracy of 96.3 per cent and an F1 score of 0.95, a measure that balances precision and recall. The system could also successfully classify eight breast cancer subtypes, with accuracy remaining above 90 per cent across all categories.
The approach combines two forms of AI. A Vision Transformer (ViT), a deep-learning model that identifies patterns across entire images, extracts features from biopsy slides. A fully connected neural network (FCNN) analyses molecular marker data. The resulting information is combined to give a clearer diagnosis.
The team says the method improves on many existing AI systems, which usually focus on image analysis and overlook molecular information that influences tumour behaviour and treatment response. The model also incorporates clinical data regarding survival trends and so can help support treatment decisions.
Zhang, Y., Zhang, Y., Xu, H. and Wang, Y. (2026) ‘Establishment of artificial intelligence pathological feature diagnosis model and molecular mechanism’, Int. J. Data Mining and Bioinformatics, Vol. 30, No. 6, pp.1–20.
