Technologies

29 pages, 2665 KiB

Open AccessReview

Data-Driven Learning Models for Internet of Things Security: Emerging Trends, Applications, Challenges and Future Directions

by Oyeniyi Akeem Alimi

Technologies 2025, 13(5), 176; https://doi.org/10.3390/technologies13050176 - 29 Apr 2025

Abstract

The prospect of integrating every object under a unified infrastructure, which provides humans with the possibility to monitor, access, and control objects and systems, has played a significant role in the geometric growth of the Internet of Things (IoT) paradigm, across various applications. [...] Read more.

The prospect of integrating every object under a unified infrastructure, which provides humans with the possibility to monitor, access, and control objects and systems, has played a significant role in the geometric growth of the Internet of Things (IoT) paradigm, across various applications. However, despite the numerous possibilities that the IoT paradigm offers, security and privacy within and between the different interconnected devices and systems are integral to the long-term growth of IoT networks. Various sophisticated intrusions and attack variants have continued to plague the sustainability of IoT technologies and networks. Thus, effective methodologies for the prompt identification, detection, and mitigation of these menaces are priorities for stakeholders. Recently, data-driven artificial intelligence (AI) models have been considered effective in numerous applications. Hence, in recent literature studies, various single and ensemble AI subset models, such as deep learning and reinforcement learning models, have been proposed, resulting in effective decision-making for the secured operation of IoT networks. Considering the growth trends, this study presents a critical review of recently published articles whereby learning models were proposed for IoT security analysis. The aim is to highlight emerging IoT security issues, current conventional strategies, methodology procedures, achievements, and also, importantly, the limitations and research gaps identified in those specific IoT security analysis studies. By doing so, this study provides a research-based resource for scholars researching IoT and general industrial control systems security. Finally, some research gaps, as well as directions for future studies, are discussed. Full article

(This article belongs to the Special Issue IoT-Enabling Technologies and Applications)

► Show Figures

Figure 1

22 pages, 561 KiB

Open AccessArticle

Opinion Mining and Analysis Using Hybrid Deep Neural Networks

by Adel Hidri, Suleiman Ali Alsaif, Muteeb Alahmari, Eman AlShehri and Minyar Sassi Hidri

Technologies 2025, 13(5), 175; https://doi.org/10.3390/technologies13050175 - 28 Apr 2025

Viewed by 41

Abstract

Understanding customer attitudes has become a critical component of decision-making due to the growing influence of social media and e-commerce. Text-based opinions are the most structured, hence playing an important role in sentiment analysis. Most of the existing methods, which include lexicon-based approaches [...] Read more.

Understanding customer attitudes has become a critical component of decision-making due to the growing influence of social media and e-commerce. Text-based opinions are the most structured, hence playing an important role in sentiment analysis. Most of the existing methods, which include lexicon-based approaches and traditional machine learning techniques, are insufficient for handling contextual nuances and scalability. While the latter has limitations in model performance and generalization, deep learning (DL) has achieved improvement, especially on semantic relationship capturing with recurrent neural networks (RNNs) and convolutional neural networks (CNNs). The aim of the study is to enhance opinion mining by introducing a hybrid deep neural network model that combines a bidirectional gated recurrent unit (BGRU) and long short-term memory (LSTM) layers to improve sentiment analysis, particularly addressing challenges such as contextual nuance, scalability, and class imbalance. To substantiate the efficacy of the proposed model, we conducted comprehensive experiments utilizing benchmark datasets, encompassing IMDB movie critiques and Amazon product evaluations. The introduced hybrid BGRU-LSTM (HBGRU-LSTM) architecture attained a testing accuracy of 95%, exceeding the performance of traditional DL frameworks such as LSTM (93.06%), CNN+LSTM (93.31%), and GRU+LSTM (92.20%). Moreover, our model exhibited a noteworthy enhancement in recall for negative sentiments, escalating from 86% (unbalanced dataset) to 96% (balanced dataset), thereby ensuring a more equitable and just sentiment classification. Furthermore, the model diminished misclassification loss from 20.24% for unbalanced to 13.3% for balanced dataset, signifying enhanced generalization and resilience. Full article

(This article belongs to the Section Information and Communication Technologies)

► Show Figures

Figure 1

31 pages, 13407 KiB

Open AccessArticle

Development of 6D Electromagnetic Actuation for Micro/Nanorobots in High Viscosity Fluids for Drug Delivery

by Maki K. Habib and Mostafa Abdelaziz

Technologies 2025, 13(5), 174; https://doi.org/10.3390/technologies13050174 - 27 Apr 2025

Viewed by 144

Abstract

This research focuses on the development, design, implementation, and testing (with complete hardware and software integration) of a 6D Electromagnetic Actuation (EMA) system for the precise control and navigation of micro/nanorobots (MNRs) in high-viscosity fluids, addressing critical challenges in targeted drug delivery within [...] Read more.

This research focuses on the development, design, implementation, and testing (with complete hardware and software integration) of a 6D Electromagnetic Actuation (EMA) system for the precise control and navigation of micro/nanorobots (MNRs) in high-viscosity fluids, addressing critical challenges in targeted drug delivery within complex biological environments, such as blood vessels. The primary objective is to overcome limitations in the actuation efficiency, trajectory stability, and accurate path-tracking of MNRs. The EMA system utilizes three controllable orthogonal pairs of Helmholtz coils to generate uniform magnetic fields, which magnetize and steer MNRs in 3D for orientation. Another three controllable orthogonal pairs of Helmholtz coils generate uniform magnetic fields for the precise 3D orientation and steering of MNRs. Additionally, three orthogonal pairs of Maxwell coils generate uniform magnetic field gradients, enabling efficient propulsion in dynamic 3D fluidic environments in real time. This hardware configuration is complemented by three high-resolution digital microscopes that provide real-time visual feedback, enable the dynamic tracking of MNRs, and facilitate an effective closed-loop control mechanism. The implemented closed-loop control technique aimed to enhance trajectory accuracy, minimize deviations, and ensure the stable movement of MNRs along predefined paths. The system’s functionality, operation, and performance were tested and verified through various experiments, focusing on hardware, software integration, and the control algorithm. The experimental results show the developed system’s ability to activate MNRs of different sizes (1 mm and 0.5 mm) along selected desired trajectories. Additionally, the EMA system can stably position the MNR at any point within the 3D fluidic environment, effectively counteracting gravitational forces while adhering to established safety standards for electromagnetic exposure to ensure biocompatibility and regulatory compliance. Full article

(This article belongs to the Special Issue IoT-Enabling Technologies and Applications)

► Show Figures

Figure 1

37 pages, 3382 KiB

Open AccessArticle

Multi-Domain Feature Incorporation of Lightweight Convolutional Neural Networks and Handcrafted Features for Lung and Colon Cancer Diagnosis

by Omneya Attallah

Technologies 2025, 13(5), 173; https://doi.org/10.3390/technologies13050173 - 25 Apr 2025

Viewed by 95

Abstract

This study presents a computer-aided diagnostic (CAD) framework that integrates multi-domain features through a hybrid methodology. The system uses several light deep networks (EfficientNetB0, MobileNet, and ResNet-18), which feature fewer layers and parameters, unlike traditional systems that depend on a single, parameter-complex deep [...] Read more.

This study presents a computer-aided diagnostic (CAD) framework that integrates multi-domain features through a hybrid methodology. The system uses several light deep networks (EfficientNetB0, MobileNet, and ResNet-18), which feature fewer layers and parameters, unlike traditional systems that depend on a single, parameter-complex deep network. Additionally, it employs several handcrafted feature extraction techniques. It systematically assesses the diagnostic power of deep features only, handcrafted features alone, and both deep and handcrafted features combined. Furthermore, it examines the influence of combining deep features from multiple CNNs with distinct handcrafted features on diagnostic accuracy, providing insights into the effectiveness of this hybrid approach for classifying lung and colon cancer. To achieve this, the proposed CAD employs non-negative matrix factorization for lowering the dimension of the spatial deep feature sets. In addition, these deep features obtained from each network are distinctly integrated with handcrafted features sourced from temporal statistical attributes and texture-based techniques, including gray-level co-occurrence matrix and local binary patterns. Moreover, the CAD integrates the deep attributes of the three deep networks with the handcrafted attributes. It also applies feature selection based on minimum redundancy maximum relevance to the integrated deep and handcrafted features, guaranteeing optimal computational efficiency and high diagnostic accuracy. The results indicated that the suggested CAD system attained remarkable accuracy, reaching 99.7% using multi-modal features. The suggested methodology, when compared to present CAD systems, either surpassed or was closely aligned with state-of-the-art methods. These findings highlight the efficacy of incorporating multi-domain attributes of numerous lightweight deep learning architectures and multiple handcrafted features. Full article

(This article belongs to the Special Issue Breakthroughs in Bioinformatics and Biomedical Engineering)

► Show Figures

Figure 1

43 pages, 15420 KiB

Open AccessReview

Advanced Precision Cutting Titanium Alloy Methods: A Critical Review Considering Cost, Efficiency, and Quality

by Guangping Wang, Xiaoxuan Chen, Zhipeng Xu, Feng Feng, Jianfu Zhang, Xiangyu Zhang and Pingfa Feng

Technologies 2025, 13(5), 172; https://doi.org/10.3390/technologies13050172 - 25 Apr 2025

Viewed by 80

Abstract

This literature review focuses on titanium alloys, which are crucial in modern manufacturing due to their excellent properties. The review covers their classification, machining challenges, and advanced cutting methods. Different alloy types (α-Ti, β-Ti, and α+β-Ti) have distinct characteristics and applications; their machining [...] Read more.

This literature review focuses on titanium alloys, which are crucial in modern manufacturing due to their excellent properties. The review covers their classification, machining challenges, and advanced cutting methods. Different alloy types (α-Ti, β-Ti, and α+β-Ti) have distinct characteristics and applications; their machining challenges include low thermal conductivity and pronounced chemical reactivity. Nowadays, advanced cutting methods of titanium alloys involve innovated tool design, efficient coolant techniques, and ultrasonic vibration cutting. The impact of these methods on cost, quality, and efficiency is analyzed, considering both positive and negative aspects. Lastly, strategies for cost reduction, efficiency improvement, and quality enhancement are explored, highlighting the complex relationship between these factors in titanium alloy processing. Full article

(This article belongs to the Collection Review Papers Collection for Advanced Technologies)

► Show Figures

Graphical abstract

37 pages, 10123 KiB

Open AccessArticle

A Novel Three-Dimensional Sliding Pursuit Guidance and Control of Surface-to-Air Missiles

by Belkacem Bekhiti, George F. Fragulis, Mohamed Rahmouni and Kamel Hariche

Technologies 2025, 13(5), 171; https://doi.org/10.3390/technologies13050171 - 24 Apr 2025

Viewed by 106

Abstract

In recent decades, missile guidance and control have advanced significantly, with methods like pure pursuit (PP), command to line-of-sight (CLOS), and proportional navigation (PN) enabling accurate target interception in uncertain environments through line-of-sight (LOS) tracking. In this work, we propose a novel 3D [...] Read more.

In recent decades, missile guidance and control have advanced significantly, with methods like pure pursuit (PP), command to line-of-sight (CLOS), and proportional navigation (PN) enabling accurate target interception in uncertain environments through line-of-sight (LOS) tracking. In this work, we propose a novel 3D sliding pure pursuit guidance (3DSPP) law for controlling a surface-to-air missile against a maneuvering target. The algorithm is compared with established guidance laws such as zero-effort miss distance “ZEM-PN” and “3D-PP”, with performance metrics including the miss distance

M_{d}

and time of closest approach

t_{c a p}

. The results demonstrate that the 3DSPP outperforms the conventional methods by achieving the lowest

M_{d} =

0.1497 m and the fastest

t_{c a p} =

7.3853 s, ensuring more precise and rapid interception. The algorithm also exhibits superior robustness to noise and efficient energy management, making it a promising solution for real-world missile guidance systems. Full article

► Show Figures

Figure 1

48 pages, 6422 KiB

Open AccessReview

Modern Trends and Recent Applications of Hyperspectral Imaging: A Review

by Ming-Fang Cheng, Arvind Mukundan, Riya Karmakar, Muhamed Adil Edavana Valappil, Jumana Jouhar and Hsiang-Chen Wang

Technologies 2025, 13(5), 170; https://doi.org/10.3390/technologies13050170 - 23 Apr 2025

Viewed by 308

Abstract

Hyperspectral imaging (HSI) is an advanced imaging technique that captures detailed spectral information across multiple fields. This review explores its applications in counterfeit detection, remote sensing, agriculture, medical imaging, cancer detection, environmental monitoring, mining, mineralogy, and food processing, specifically highlighting significant achievements from [...] Read more.

Hyperspectral imaging (HSI) is an advanced imaging technique that captures detailed spectral information across multiple fields. This review explores its applications in counterfeit detection, remote sensing, agriculture, medical imaging, cancer detection, environmental monitoring, mining, mineralogy, and food processing, specifically highlighting significant achievements from the past five years, providing a timely update across several fields. It also presents a cross-disciplinary classification framework to systematically categorize applications in medical, agriculture, environment, and industry. In counterfeit detection, HSI identified fake currency with high accuracy in the 400–500 nm range and achieved a 99.03% F1-score for counterfeit alcohol detection. Remote sensing applications include hyperspectral satellites, which improve forest classification accuracy by 50%, and soil organic matter, with the prediction reaching R² = 0.6. In agriculture, the HSI-TransUNet model achieved 86.05% accuracy for crop classification, and disease detection reached 98.09% accuracy. Medical imaging benefits from HSI’s non-invasive diagnostics, distinguishing skin cancer with 87% sensitivity and 88% specificity. In cancer detection, colorectal cancer identification reached 86% sensitivity and 95% specificity. Environmental applications include PM2.5 pollution detection with 85.93% accuracy and marine plastic waste detection with 70–80% accuracy. In food processing, egg freshness prediction achieved R² = 91%, and pine nut classification reached 100% accuracy. Despite its advantages, HSI faces challenges like high costs and complex data processing. Advances in artificial intelligence and miniaturization are expected to improve accessibility and real-time applications. Future advancements are anticipated to concentrate on the integration of deep learning models for automated feature extraction and decision-making in hyperspectral imaging analysis. The development of lightweight, portable HSI devices will enable more on-site applications in agriculture, healthcare, and environmental monitoring. Moreover, real-time processing methods will enhance efficiency for field deployment. These improvements seek to enhance the accessibility, practicality, and efficacy of HSI in both industrial and clinical environments. Full article

(This article belongs to the Special Issue Artificial Intelligence and Smart Information Systems: Trends and Innovations)

► Show Figures

Figure 1

23 pages, 5095 KiB

Open AccessArticle

Human-Machine Interaction: A Vision-Based Approach for Controlling a Robotic Hand Through Human Hand Movements

by Gerardo García-Gil, Gabriela del Carmen López-Armas and José de Jesús Navarro, Jr.

Technologies 2025, 13(5), 169; https://doi.org/10.3390/technologies13050169 - 23 Apr 2025

Viewed by 113

Abstract

An anthropomorphic robot is a mechanical device designed to perform human-like tasks, such as manipulating objects, and has been one of the significant contributions in robotics over the past 60 years. This paper presents an advanced system for controlling a robotic arm using [...] Read more.

An anthropomorphic robot is a mechanical device designed to perform human-like tasks, such as manipulating objects, and has been one of the significant contributions in robotics over the past 60 years. This paper presents an advanced system for controlling a robotic arm using user hand gestures and movements. It eliminates the need for traditional sensors or physical controls by implementing an intuitive approach based on MediaPipe and computer vision. The system recognizes the user’s hand movements. It translates them into commands that are sent to a microcontroller, which operates a robotic hand equipped with six servomotors: five for the fingers and one for the wrist, which stands out for its orthonormal design that avoids occlusion problems in turns of up to 180°, guaranteeing precise wrist control. Unlike conventional systems, this approach uses only a 2D camera to capture movements, simplifying design and reducing costs. The proposed system allows replicating the user’s activity with high precision, expanding the possibilities of human-robot interaction. Notably, the system has been able to replicate the user’s hand gestures with an accuracy of up to 95%. Full article

(This article belongs to the Special Issue Image Analysis and Processing)

► Show Figures

Graphical abstract

30 pages, 13157 KiB

Open AccessArticle

Development of IoT-Based Hybrid Autonomous Networked Robots

by Maki K. Habib and Chimsom I. Chukwuemeka

Technologies 2025, 13(5), 168; https://doi.org/10.3390/technologies13050168 - 23 Apr 2025

Viewed by 117

Abstract

Autonomous Networked Robot (ANR) systems feature multi-robot systems (MRSs) and wireless sensor networks (WSNs). These systems help to extend coverage, maximize efficiency in data routing, and provide practical and reliable task management, among others. This article presents the development and implementation of an [...] Read more.

Autonomous Networked Robot (ANR) systems feature multi-robot systems (MRSs) and wireless sensor networks (WSNs). These systems help to extend coverage, maximize efficiency in data routing, and provide practical and reliable task management, among others. This article presents the development and implementation of an IoT-based hybrid ANR system integrated with different cloud platforms. The system comprises two main components: the physical hybrid ANR, the simulation development environment (SDE) with hardware in the loop (HIL), and the necessary core interfaces. Both are integrated to facilitate system component development, simulation, testing, monitoring, and validation. The operational environment (local and/or distributed) of the designed system is divided into zones, and each zone comprises static IoT-based sensor nodes (SSNs) and a mobile robot with integrated onboard IoT-based sensor nodes (O-SSNs) called the mobile robot sensor node (MRSN). Global MRSNs (G-MRSNs) navigate spaces not covered by a zone. The mobile robots navigate within/around their designated spaces and to any of their SSNs. The SSNs and the O-SSN of each zone are supported by the ZigBee protocol, forming a WSN. The MRSNs and G-MRSNs communicate their collected data from different zones to the base station (BS) through the IoT base station gateway (IoT-BSG) using wireless serial protocol. The base station analyzes and visualizes the received data through GUIs and communicates data through the IoT/cloud using the Wi-Fi protocol. The developed system is demonstrated for event detection and surveillance. Experimental results of the implemented/simulated ANR system and HIL experiments validate the performance of the developed IoT-based hybrid architecture. Full article

(This article belongs to the Special Issue IoT-Enabling Technologies and Applications)

► Show Figures

Figure 1

20 pages, 3149 KiB

Open AccessReview

The Landscape of Virtual Reality Use in Mobility Rehabilitation from 2010–2023: A Scoping Review

by Danielle T. Felsberg, Reza Pousti, Charlend K. Howard, Scott E. Ross, Louisa D. Raisbeck, Jared T. McGuirt and Christopher K. Rhea

Technologies 2025, 13(5), 167; https://doi.org/10.3390/technologies13050167 - 22 Apr 2025

Viewed by 259

Abstract

Significant advancements in virtual reality (VR) technology have occurred in the past decade, allowing clinical researchers to take advantage of these reduced barriers to explore the use of VR in patient populations. This scoping review on VR interventions to improve mobility in adults [...] Read more.

Significant advancements in virtual reality (VR) technology have occurred in the past decade, allowing clinical researchers to take advantage of these reduced barriers to explore the use of VR in patient populations. This scoping review on VR interventions to improve mobility in adults and children focuses on the literature from 2010–2023. A total of 2736 articles were screened and 126 articles met the inclusion criteria. Most of the studies were conducted in inpatient clinical settings (n = 41) and investigated VR interventions to improve balance (n = 118). Less immersive (n = 108) products such as Nintendo Wii or Xbox Kinect were primarily used. Additionally, 37.0% of studies (n = 47) used off-the-shelf programs like Wii Fit Plus and 73.2% of studies (n = 93) found statistically significant improvements in motor outcomes following VR intervention. The articles included in this review suggest that the majority of VR research for physical rehabilitation is being performed in clinical settings. Most studies reported statistically significant improvements in their outcome variables following VR intervention. These observations demonstrate that research in this area is moving beyond proof-of-concept and toward translation to clinical applications. Full article

(This article belongs to the Special Issue Assistive Technologies in Care and Rehabilitation: Research, Developments, and International Initiatives)

► Show Figures

Figure 1

26 pages, 3862 KiB

Open AccessArticle

Application of a Hybrid Model for Data Analysis in Hydroponic Systems

by Kuanysh Bakirov, Jamalbek Tussupov, Akhmet Tussupov, Ibraheem Shayea and Aruzhan Shoman

Technologies 2025, 13(5), 166; https://doi.org/10.3390/technologies13050166 - 22 Apr 2025

Viewed by 346

Abstract

This study presents a hybrid data analysis approach to optimize the growing conditions for beetroot and tarragon microgreens cultivated in hydroponic systems. Maintaining precise microclimate control is essential, as even minor deviations can significantly affect the yield and product quality, but traditional monitoring [...] Read more.

This study presents a hybrid data analysis approach to optimize the growing conditions for beetroot and tarragon microgreens cultivated in hydroponic systems. Maintaining precise microclimate control is essential, as even minor deviations can significantly affect the yield and product quality, but traditional monitoring methods fail to adapt promptly to changing conditions. To overcome this limitation, an automated monitoring system integrating machine learning methods XGBoost 3.0.0, principal component analysis (PCA), and fuzzy logic was developed. The model continuously identifies the deviations in environmental parameters and recommends corrective actions to stabilize the growth conditions. Experimental evaluation demonstrated superior predictive performance by using XGBoost, achieving an accuracy and F1-score of 97.88%, ROC-AUC of 99.99%, and computational efficiency (training completed in 2.3 s), outperforming RandomForest and GradientBoosting algorithms. Real-time data collection was facilitated through IoT sensors transmitting readings via Wi-Fi every 5 s to a local server, accumulating approximately 17,280 records per day. The analysis highlighted air humidity, solution humidity, and temperature as critical influencing factors. This research confirms the developed system’s effectiveness in intelligent hydroponic monitoring, with future work aimed at integrating IoT and IIoT technologies for scalable management across diverse crops. Full article

(This article belongs to the Section Information and Communication Technologies)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Technologies, Volume 13, Issue 5 (May 2025) – 11 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI