Volume 21 Issue 5 2026

Background and Objective Sickle Cell Disease is a blood genetic disorder and a major problem among the tribal population of India. There is no structured program in action in the country for the diagnosis, treatment, and management of the disease. This study protocol aims to train health care workers i.e., Medical officers, Auxiliary Nursing Midwifery , Laboratory Technician, and Community Health Officers , for diagnosis and counseling for sickle cell disease through a training program in eight SCD- endemic tribal districts of Rajasthan, India. Methodology The sub-districts of Udaipur, Banswara, Sirohi, Dungarpur, Pratapgarh, Pali, Chittorgarh, and 25 Rajsamand will be included in the study. The sub-districts having more than 50% tribal 26 population will be included in the study. The training sessions will be organized at the CHC/PHC in eight districts. Data for this study will be collected from pre- and post-questionnaires given to healthcare professionals during the training program. A counseling module in English and the 29 local language will be circulated to health worker to improve their knowledge regarding 30 diagnosis, management, and prevention strategies for sickle cell disease. Discussion The results of this study could provide information on the necessity of bolstering the capacity for implementation research in endemic areas. Expected outcomes The outcomes of the study will provide a better understanding regarding the diagnosis, control, and management of sickle cell disease among the health workers. The primary outcome will be capacity building of the health workers in conducting screening of SCD and creating awareness of sickle cell disease among tribes.

The rapid expansion of Nigeria's fintech ecosystem, accelerated by recent cashless policy drives, presents a transformative potential for the nation's economic landscape. However, the informal sector, which constitutes a significant portion of Nigeria's GDP, remains largely marginalized in the digital financial narrative. This phenomenological study explores the lived experiences and perceptions of informal sector Small and Medium Enterprises (SMEs) regarding their adoption and utilization of fintech solutions. Through in-depth interviews with traders and artisans, the research reveals a complex interplay of trust, convenience, and systemic barriers. While participants appreciate the efficiency and security of digital payments over physical cash, they express profound apprehension regarding transaction failures, hidden charges, and the prevalence of electronic fraud. The study finds that utilization is largely reactive rather than strategic; SMEs adopt tools like USSD and payment gateways primarily to meet customer demand rather than to optimize their own business operations. The findings suggest that current fintech interfaces often fail to accommodate the low-digital literacy and high-risk aversion characteristic of the informal sector. Consequently, the study recommends a paradigm shift towards user-centric designs, robust consumer protection frameworks, and targeted digital literacy programs to bridge the trust gap and integrate informal SMEs effectively into the formal digital economy.

The removal of the Premium Motor Spirit (PMS) subsidy in Nigeria represents a pivotal macroeconomic shock that drastically escalated operational costs, threatening the survival of the non-oil sector. This study investigates the resilience of Small and Medium Enterprises (SMEs) by exploring the adaptive strategies and coping mechanisms these businesses have employed in the wake of this policy reform. Adopting a multiple case study design, the research utilizes in-depth semi-structured interviews and observational data from 20 SMEs across the manufacturing, retail, and service sectors. The analysis reveals a complex spectrum of survival strategies, ranging from immediate cost-cutting measures such as staff downsizing, inventory rationing, and reduction of operating hours to more structural adaptations, including the aggressive passing of costs to consumers and the tentative shift toward alternative energy sources. The findings indicate that while SMEs exhibit significant organizational resilience, these coping mechanisms are predominantly reactive and increasingly unsustainable, pushing many firms to the brink of insolvency. The study concludes that ensuring the longevity of the SME sector requires government interventions that move beyond general palliatives to include targeted fiscal relief, subsidized energy tariffs, and access to low-interest credit facilities tailored to the post-subsidy economic reality.

The study examined the relationship between selected Artificial Intelligence (AI) tools and the management of teaching and learning in Universities in the North-Central States of Nigeria. Specifically, the study focused on the use of chatbots and automated grading systems. A correlational design was adopted, and a sample of 386 university staff was drawn using stratified random sampling from a population of 10,787 staff from nine (9) Universities in North-central Nigeria, determined by the Taro Yamane calculator. Data was collected using a 16-item questionnaire titled Artificial Intelligence and the management of teaching and learning in Universities in the North-Central States of Nigeria (AIMTLUNCN). The instrument was validated by experts, yielding a validity index of 0.79, while the reliability coefficient of 0.71 was obtained through pilot testing. Pearson’s Product-Moment Correlation was used to answer the research questions, while Regression analysis was used to test the hypotheses of the study at a 0.05 level of significance. Findings reveal a very weak and non-significant relationship between Chatbots and the management of teaching and learning (r = 0.062, p > 0.05) as well as between automated grading systems and the management of teaching and learning (r = 0.026, p > 0.05). The study concludes that there is no significant relationship between the selected AI tools and the management of teaching and learning, indicating limited adoption and ineffective integration of AI in university teaching and learning management systems.

This study investigated the relationship between Artificial intelligence-based adaptive learning systems, intelligent tutoring systems and students’ learning in universities in North-central Nigeria. The study was guided by two (2) research objectives, questions, and corresponding hypotheses. The study was anchored on the Technology Acceptance Model (TAM) and adopted a correlational research design. The population comprised 10,787 students across nine universities in North-Central, from which a sample size of 399 was selected using a stratified random sampling. Data was collected using a validated questionnaire and analysed through Pearson’s Product-Moment Correlation, and Regression analysis was used to test the formulated hypotheses at a 0.05 level of significance. The finding revealed that both the adaptive learning systems and the intelligent tutoring systems have a weak positive relationship with students’ learning (r=0.055) and were not statistically significant. Consequently, both hypotheses were accepted. The result indicated that although AI technologies have the potential to enhance personalised and interactive learning, their current utilisation in universities within the North-central zone of Nigeria has not significantly influenced students’ learning outcomes. This weak relationship suggests low acceptance and utilisation of AI tools, which aligns with TAM constructs of perceived usefulness and perceived ease of use. The study concludes that the limited impact of these technologies may be linked to low perceived usefulness and ease of use, as well as infrastructural and institutional challenges affecting their adoption. It is recommended that universities should improve technological infrastructure, enhance digital literacy and provide adequate support systems to facilitate the effective integration of AI tools in teaching and learning

This research article explores the origins, historical significance, ritual practices, and sociocultural importance of Maa Abhaya, a local folk deity worshipped in Kutasingha village, Balangir district, Odisha, India. The veneration of Maa Abhaya is deeply rooted in the collective memory of the community and is linked to a miraculous healing event during an ancient cholera epidemic. Notably, the cult distinguishes itself within the landscape of Odishan folk religion by its unwavering commitment to non-violent offerings primarily coconuts in sharp contrast to the widespread practice of animal sacrifice among village deities in the region. The annual festival, celebrated during the auspicious month of Jyeshtha (Jyoshtha Masa), serves as a significant unifying occasion for the community. Drawing from oral traditions, ethnographic observation, and comparative religious analysis, this article contends that the cult of Maa Abhaya represents a rare embodiment of ahimsa (non-violence) within an ancient folk religious context. This unique tradition merits scholarly attention and cultural preservation for its exceptional commitment to non-violence and its enduring role in local religious identity.

Background: Radiation exposure remains a critical concern in pediatric computed tomography (CT) due to heightened radiosensitivity and increased lifetime attributable risk of malignancy. Technological advances in image reconstruction have enabled substantial radiation dose optimization. Deep learning reconstruction (DLR), a novel artificial intelligence-based approach, has emerged as a promising method to improve image quality while permitting further dose reduction. Purpose: To systematically evaluate current evidence regarding the impact of deep learning reconstruction on image quality and radiation dose reduction in pediatric low-dose CT. Method: A systematic review was conducted in accordance with PRISMA 2020 guidelines. PubMed and Google Scholar were searched for English-language original research article published between January 2019 and March 2026. Studies were eligible if they include pediatric patients (≤18 years), evaluated CT imaging with deep learning reconstruction and reported objective or subjective image quality metrics and/or radiation dose outcomes. Data were extracted using standardized forms, and risk of bias was assessed using an adapted Newcastle-Ottawa framework. Due to heterogeneity in reporting and incomplete availability of variance data, a structured narrative synthesis was performed. Result: Fifty-three records were identified; for studies met inclusion criteria. All included studies demonstrated significant reductions in image noise and improvements in signal-tonoise ratio (SNR) and contrast-to-noise ratio (CNR) with DLR compared to conventional iterative reconstruction techniques. Two studies reported radiation dose reductions of approximately 50-55% while maintaining diagnostic image quality. Risk of bias was low to moderate across studies. Conclusion: Available pediatric evidence indicates that deep learning reconstruction improves image quality and enables meaningful radiation dose reduction in low- dose CT protocols. Although current data remain limited, findings consistently support integration of DLR into pediatric CT imaging strategies.

Figurative language (metaphors, idioms, proverbs, evaluative comparisons, and symbolic images) is one of the most sensitive zones in which a society’s gender concept becomes visible. Gender, understood as a socio-cultural construct rather than a purely biological category, is repeatedly encoded in imagery that evaluates women and men, prescribes “appropriate” behavior, and legitimizes power relations through seemingly neutral expressions [1, 2]. This article examines how the gender concept operates in figurative language as (1) a cognitive mechanism that maps abstract social meanings onto concrete bodily and cultural images [3], (2) a discourse mechanism that indexes stereotypes and normative expectations [4, 5], and (3) a pragmatic mechanism that strengthens persuasion and social control in everyday interaction and media. Using a qualitative comparative approach, the study analyzes representative figurative patterns from Azerbaijani, Russian, English, and selected European languages (French, German, Italian), with brief parallels from Turkish context, to demonstrate cross-cultural similarities (e.g., metaphors of strength, purity, danger, leadership) and culturally specific models (e.g., honor/shame frames; family-role symbolism). Brief parallels from Turkish figurative language further support the cross-cultural patterns discussed above. In Turkish, masculinity is frequently conceptualized through metaphors of moral strength and responsibility, as in “Erkek adam sözünün eri olur” (“A real man stands by his word”) or “aslan gibi adam” (“a man like a lion”), which align leadership and authority with strength and reliability. Femininity, by contrast, is often evaluated through metaphors of modesty, honor, and domestic responsibility, for example “Yuvayı dişi kuş yapar” (“The female bird builds the nest”) or “namus kadının süsüdür” (“Honor is a woman’s ornament”). These Turkish examples parallel Azerbaijani and Russian data by showing how figurative language encodes gender roles through culturally salient values such as honor, family reputation, and moral restraint, rather than through explicit gender labeling.The discussion integrates conceptual metaphor theory, gender linguistics, and critical discourse perspectives to show that gendered imagery is not decorative: it structures meaning, guides interpretation, and shapes attitudes in communication. The article proposes a classroom- and research-friendly analytical framework for identifying gender metaphors and gendered idioms, with practical implications for language education, translation, and intercultural communication—especially in contexts such as Azerbaijan where multilingual contact intensifies the circulation of global and local figurative models [6].

A comprehensive investigation into the progress in solar water heating technologies since inception is presented in this paper. The efficient utilization of solar energy for heating water is crucial for sustainable energy practices, and thermal energy storage in the form of latent heat emerges as a promising solution. The research summarizes various studies examining thermal storage systems with and without phase change materials (PCMs), classifying them into the kind of collector and the category of storage in action (either sensible or latent). An exhaustive literature study underscores the importance of PCM selection criteria, emphasizing the requirement for materials that possess high latent heat capacity and extensive surface areas to ensure heat transfer optimally. This article highlights the significance of PCM-based thermal storage in enhancing the performance (thermal) of solar water heaters, paving the way for more efficient and sustainable heating solutions.

Centrifugal fans are used extensively in industries where their aerodynamics play an important role in affecting the efficiency of the process and energy consumed. This paper reviews the state of the art related to centrifugal fan design, aerodynamics, CFD analyses, optimization methods, experimentation, structure, and noise. Analysis of the effect of important geometric variables, such as blades profile, impeller arrangement, and volute shape, is done to determine their impact on performance. Effectiveness of numerical models, especially Computational Fluid Dynamics (CFD), in simulating flow, pressure, and losses in the fan is evaluated. Several optimization methods, such as Response Surface Methodology (RSM), Genetic Algorithm (GA), Artificial Neural Network (ANN), and multi-objective optimization, are discussed. The significance of experiments and structural analysis for reliable and durable systems is also explored. The key trends, research areas, and recommendations have been developed based on the comparative analysis of the existing literature to provide insights into the design of efficient centrifugal fans.

The article examines the nature of humanity through a synthesis of psychological theories and contemporary data from paleogenetics. Classical psychological theories do not account for the genetic diversity of modern humans. Meanwhile, modern achievements in paleogenetics pose the problem of revising many models of human behavior and nature. A theoretical, comparative analysis of works by classics of psychology and articles from the search systems PubMed, Google Scholar, and e-library was conducted, as well as a descriptive analysis of modern paleogenetic studies on the sequencing of hominin genes. The research hypothesis suggests that differences are conditioned not only by social factors but also by the biological heritage from various hominin species. The study's main results are presented in a critique of Maslow's concept of the hierarchy of needs, which cannot be viewed as universal. The theory ignores the biological diversity of humans, who have inherited up to 6% of their genes from Neanderthals and Denisovans, which may influence behavior. Biological heritage explains the variability in human needs. Cultural differences also alter the "hierarchy" of humanity. It is shown that humanity cannot be reduced to a single model, as it depends on genetics, culture, and context. It is emphasized that the evolutionary approach explains such manifestations in different groups of people as aggression, creativity, or altruism. However, while there is no direct evidence of a link between specific genes and psychological traits, the hypothesis requires further research. Humanity is the product of a complex interaction between biology (genes of ancient hominins) and the social environment. Psychological theories must integrate paleogenetic data for more accurate models of motivation. Interdisciplinary research at the intersection of psychology, anthropology, and genetics can yield new analytical material.

The present study is aimed to evaluate the antidiabetic and antioxidant properties of a newly synthesized Metformin–Resveratrol Aldehyde complex (Met-Res-Aldehyde complex) in high-fat diet-fed, lowdose streptozotocin-induced experimental type 2 diabetes mellitus in rats. The effect of oral administration of the Met-Res-Aldehyde complex (5 mg/kg body weight) for a period of 30 days on the levels of biochemical parameters was evaluated in experimental groups of rats. The antidiabetic efficacy of the complex were assessed by measuring a range of biochemical indices, including fasting blood glucose, plasma insulin, haemoglobin, glycosylated haemoglobin, total protein, urea, uric acid, and creatinine. Oxidative stress markers such as TBARS, lipid peroxides, hydroperoxides and protein carbonyls were analyzed in plasma, pancreatic, hepatic, and renal tissues. The status of enzymatic antioxidants, including superoxide dismutase (SOD), catalase, glutathione peroxidase (GPx), and glutathione S-transferase (GST), gluatathione reductase (GR) as well as non-enzymatic antioxidants such as vitamin E, vitamin C and ceruloplasmin, were also evaluated. Diabetic rats showed significantly increased levels of fasting blood glucose and glycosylated haemoglobin. Oral treatment with the Met-Res-Aldehyde complex resulted in the maintenance of normoglycemia by decreasing oxidative stress markers and improving antioxidant status in diabetic rats. The results of the study indicate that the Met-Res-Aldehyde complex is non-toxic and possesses significant antioxidant properties which in turn responsible for its observed antidiabetic efficacy. These effects are comparable to those of metformin, a standard oral hypoglycemic drug.

Advancing smart agricultural infrastructure requires intelligent systems that can sense, simulate, and optimize physical assets across their life cycle. Digital twin frameworks provide a virtual representation of fields, water networks, machinery, and storage facilities, enabling continuous data-driven monitoring, predictive maintenance, and scenario-based planning. By integrating Internet of Things sensing, geospatial analytics, and artificial intelligence, digital twins can dynamically represent the structural, agronomic, and environmental status of critical farm and regional infrastructure, thereby supporting timely interventions and resilient operations. This chapter conceptualizes a layered digital twin architecture for smart agriculture that links physical infrastructure with interoperable data platforms, simulation engines, and decision-support dashboards. It highlights innovations in conditionbased and predictive maintenance, risk-aware asset planning, and resource-efficient scheduling for irrigation, energy use, and logistics. Special emphasis is placed on scalability from farm to landscape levels, interoperability with existing agricultural management systems, and the role of standards and governance in ensuring secure data exchange. Case-driven discussions demonstrate how digital twinenabled maintenance and planning can reduce downtime, extend asset lifespans, and enhance sustainability outcomes amid climate and market uncertainty. The chapter concludes by outlining research and implementation gaps, including edge analytics, domain-specific ontologies, and integration with policy and financial instruments to accelerate adoption in diverse agro-ecological contexts.

This article deals with the problems of studying communication links in a digital society. An analysis is made of the heuristic possibilities of the classical causal model for describing the dependencies of social phenomena. In the world of quantum phenomena, however, any measurement affects the system. The mere fact that we measure, for example, the location of a particle, leads to an unpredictable change in its speed. The same phenomena can be observed in the social system: any empirical study of social phenomena automatically produces in itself, some changes in this system. Modern physics has discovered the quantum structure of the material world. And human beings, as material objects, really are quantum. Therefore, social life requires a quantum structure for its proper understanding. This means that we need to develop a “quantum social science”. The principles of quantum mechanics are relevant for social theory, and therefore quantum theory cannot be ignored in the social sciences. The quantum effect can be described as the propensity of certain individuals to modify their conduct due to their consciousness of being observed. This effect refers to the fact that people will modify their behavior simply because they are being observed. The new approaches are proposed for using the quantum dependence model of theoretical physics in relation to the human phenomena in the modern digital space.

BACKGROUNG: Computed Tomography of the head is widely used in pediatric imaging due to its rapid acquisition and high diagnostic accuracy. However, concerns regarding radiation exposure in children remain significantly, as pediatric patients are more sensitive to ionizing radiation and have a longer lifetime risk of radiation-related effects. Dose Length Product is a commonly used parameter for assessing radiation dose in CT examination. Understanding trends and variability in DLP values is essential foe optimizing radiation safety in pediatric non-contrast CT head imaging. AIM: The review aimed to assess reported DLP values in pediatric patients undergoing NCCT of the head, with particular emphasis on variation across publication years, patient age and imaging protocols. METHODS: A structured review of literature published between 2020 and 2025 was conducted using major electronic databases. Studies were included if they involved paediatric patients (0-18 years) evaluated NCCT head examination and reported DLP values. Data were extracted on study characteristics, patient age range, imaging protocol and mean DLP values. A total of 80 studies met the inclusion criteria. Descriptive analysis was performed to evaluate trends in radiation dose across years, age groups, and protocol types. Standard deviation values were incorporated to assess variability in reported DLP measurements. RESULTS: The reported mean DLP values demonstrated considerable variation across studies, ranging from approximately 150 to 1000 mGy.cm, with an overall mean of 539.84 ± 132 mGY.cm. Year wise analysis showed a fluctuating pattern, with mean DLP values of 550.94 mGY.cm in 2020, increasing to 584.78 mGy.cm in 2021, peaking at 629.73 mGY.cm in 2023 and subsequently declining to 483.18 mGY.cm in 2024 and 451.58 mGY.cm in 2025. Agebased evaluation did not reveal a consistent linear relationship between age and DLP, indicating that patient age alone is not a primary determinant of radiation dose. Protocol-wise comparison showed that ultra-low-dose protocol were associated with the lowest mean DLP (507.12 mGY.cm) while standard and low-dose protocols demonstrated higher values, reflecting variability in implementation across institutions. CONCLUSION: Radiation dose in paediatric NCCT head imaging demonstrates substantial variability influenced by imaging protocols, technological factors and institutional practices. Although recent trends suggest a gradual reduction in DLP values, inconsistencies remain. The findings highlight the importance of standardized protocols and continued emphasis on dose optimization to enhance radiation safety in paediatric CT imaging.

Data science is a rapidly developing topic in the technology industry that has garnered a lot of interest recently because of its profound impact on a wide range of various disciplines or enterprises. The goal of the area is to extract meaningful perceptivity and knowledge from big, complicated databases using a variety of methods and access methods. Traditional advertising methods, characterized by broad reach and limited personalization, are giving way to targeted campaigns leveraging ML algorithms to identify and engage specific consumer segments. Targeted advertising, leveraging machine learning (ML), has emerged as a powerful tool for businesses seeking to optimize their advertising expenditure and improve return on investment (ROI). This research is concentrated and provide insurance on ad campaigns using machine learning to deliver targeted advertisements to possible clients. To make the advertisement campaign as effective as possible, machine learning algorithms would dissect data from multiple sources. By combining the analyzed data, a largely targeted advertisement campaign could be generated that delivers the right communication to the right audience at the right time. Learning opportunities from KNeighborsClassifier (KNC), Logistic Regression and Random Forest can achieve better ROI and drive more conversions with each more potential customer with their specific needs and interests. Moreover, it examines the economic impact of this paradigm shift, analyzing improvements in advertising efficiency, revenue generation, consumer surplus, and potential societal drawbacks. We argue that ML-powered targeted advertising, while presenting unique challenges, offers significant opportunities to enhance economic efficiency and deliver more value to both businesses and consumers, provided it is implemented responsibly and ethically. In addition, it explores the economic and financial impact of implementing ML-driven targeted advertising campaigns.

Background. Atrial fibrillation (AF) and dementia often coexist in elderly. It is supposed that patients with AF have higher risk for development of incident dementia. The aim of our study was to examine the impact of AF on cognitive functions and cognitive decline of patients with Alzheimer’s disease (AD), vascular cognitive impairment (VCI) and mixed (vascular and Alzheimer’s) dementia (MxD). Material and methods: We examined 146 patients with cognitive impairment (76.71±7.47 year-old, 48 males and 98 females; 38 with AD, 56 with VCI and 52 with MxD; 37 with and 109 without AF). Al patients were examined twice (in interval of 2 years) with the following neuropsychological battery: Mini Mental State Examination (MMSE), 10 Words Memory Test, 5min - delayed recall (DR) and Word Recognition Test (WRT)), Isaac Set Test (IST), Literal Fluency Test (LFT), 90-sec.-Digit Symbol Substitution Test, Digit Span Test (DST) forward and backward and Clock Drawing Test (CDT). All results were interpreted for p<0.05. Results: In total, patients with AF showed more pronounced decrease in IST performance. Among AD group, patients with AF had more severe decline in executive functioning. Among patients with VCI, patients with AF showed more severe decline in MMSE, IST, LFT, STM, DR and CDT. We failed to find any cognitive differences between patients with and without AF in MxD group.

Cantonese opera is a traditional Chinese opera genre popular in Hong Kong since the late nineteenth century. However, the genre has faced declining audience attendance in recent decades, despite promotion by Hong Kong government through various channels. There is a dearth of research on factors shaping audience motivation to attend live performances of this genre. This article reports an investigation of audience motivations to attend Cantonese opera performances using an online survey in Hong Kong. A total of 3,447 valid questionnaires were collected while MANOVA and ANOVA methods were used in the data analysis. Significant factors determining audience attendance included the intrinsic value of the performance, the quality of the artists, the quality of the performance venue, and an appropriate performance duration. While artists should strive for continuous improvement, the duration of performances might need to be shortened to suit attendees’ busy schedules.

Manual approaches of carrying out stock auditing and inventory management have been found to be inefficient due to inefficiency and susceptibility to errors and delays caused by the lack of real-time monitoring capacity. This research paper proposes the design and development of an Autonomous Mobile Robot (AMR) for warehouse inventory auditing using NVIDIA Jetson's AI technology. The system comprises sensors, cameras, and mobility units to support its autonomous navigation capacity as well as real-time monitoring of the activities taking place in the warehouse environment. The Deep Learning YOLOv8 detection model will be adopted for the automatic tracking of shelf items in the warehouse. ROS2 will be used for robot control and communication purposes, while SLAM techniques will aid localization and map-making in the process of inventory management. This research proposal outlines how this will be done through simulation tests as well as live experiments on the robot.

This paper presents an IoT based smart helmet system developed using ESP32 for rider safety, accident detection, and emergency communication. The system consists of two ESP32 modules: helmet module and bike module. A push button inside the helmet detects whether the rider is wearing the helmet properly. Only after helmet detection, the fuel pump connected to the bike module is activated, allowing vehicle operation. An MPU6050 accelerometer and gyroscope sensor continuously monitors rider head tilt and acceleration values. If the head tilt exceeds 45° and the acceleration suddenly becomes stationary, the system identifies an accident condition. During accidents, a buzzer is activated to help nearby people locate the rider. A NEO-6M GPS module provides real-time location coordinates to a dedicated MIT App Inventor mobile application which displays helmet status, live location, and emergency alerts. An SOS button is also included for women safety and emergency communication. The proposed system improves rider safety, emergency response, and intelligent vehicle control.

Another difficulty has been in the field of robotics and prosthetics, where delicate and malleable objects have been a challenge because of the difficulty in maintaining stability and force when handling them so that they don't slip and break. The presented project is related to developing an intelligent robotic hand with the help of a tactile sensor system. Two types of sensors are developed for the current work; Force Sensitive Resistors (FSRs) for static force measurement applied to grasping the objects and flex sensors for dynamics measurement of the fingers. In this way, it is possible to detect in a real-time mode sign of any possible instability in grasping. With the help of adaptive control and by changing the force and motion of fingers depending on the variation of these parameters, it is possible to stabilize the object and not allow its slipping. It was found that the implementation of multimodal tactile sensing with the use of FSRs and Flex sensors allows manipulating with fragile items in a safer way

Sixth-generation mobile networks are envisioned to be AI-native, ultra-dense, and tightly integrated with edge computing to support immersive, mission-critical, and semanticaware applications such as holographic communications, industrial digital twins, and large-scale autonomous systems [8], [13], [19], [20]. Edge computing will therefore evolve from the current 5G multi-access edge computing into a heterogeneous, intelligent, and globally distributed cloud–edge–device–non-terrestrial continuum [9], [10], [12]. While existing surveys address 6G visions, intelligent MEC, or security/privacy aspects in isolation [8], [11], [14], a holistic treatment of edge computing for 6G—covering architectures, challenges, and research opportunities—remains limited [13], [19]. This article presents a comprehensive survey of edge computing in 6G networks, starting from enabling concepts and reference architectures and then detailing the role of AI-native and semantic-aware design, integration with non-terrestrial networks, and support for diverse vertical use cases [11], [12], [15]. The survey systematically identifies open research gaps includ- ing semantic-aware edge architectures, end-to-end orchestration across heterogeneous domains, security and privacy for distributed AI models [16], energy-efficient and green edge designs, heterogeneous hardware and model management, mobility and session continuity, interoperability and standardization, benchmarking and testbeds, privacy-preserving collaborative learning with real-time constraints, and NTN-edge integration [17], [18], [19], [20]. For each gap, the article discusses the limitations of current approaches and outlines concrete research directions along with design guidelines [13], [14], [19]. The goal is to provide a unified reference for researchers and practitioners to design scalable, trustworthy, and sustainable edge systems that can realize the full potential of 6G networks [8], [20]. Index Terms—6G, edge computing, multi-access edge computing (MEC), semantic communications, AI-native networks, nonterrestrial networks (NTN), security, orchestration.