In today's fast-changing and tech-driven business world, companies have to keep changing to be strong, competitive, and ready for the future. This chapter looks at how digital transformation and artificial intelligence (AI) help organizations be more flexible by speeding up decision-making, improving forecasting powers, and increasing adaptability. It looks into how digital maturity, leadership adaptability, and technology integration methods work together to promote systemic agility. The chapter also talks about how important it is to have a workforce that is digitally educated and the policy frameworks that enable the use of technology in a way that is ethical and sustainable. The chapter gives organisations a path for finding a balance between technical progress and human-centered flexibility by combining ideas from both theoretical models and real-world examples. This will help them flourish in times of uncertainty and constant change.

Parkinson’s disease (PD) is a progressive neurological disorder characterized by motor symptoms such as tremor, rigidity, and bradykinesia. Wearable inertial sensors enable non-invasive and cost-effective assessment of motor abnormalities in real-world settings. Despite recent advances in deep learning, many existing approaches rely on complex architectures with limited interpretability and inconsistent evaluation protocols. This study proposes a simple and reproducible classical machine learning pipeline for PD detection using smartwatch-based inertial signals from the PADS dataset. Spectral and statistical features were extracted from accelerometer and gyroscope signals, and LASSObased feature selection was applied within a nested subject-level cross-validation framework to prevent data leakage. Several classifiers, including Logistic Regression, Support Vector Machine, Random Forest, CatBoost, and Multi-Layer Perceptron, were evaluated. The proposed pipeline achieved 79.26% balanced accuracy, 87.32% accuracy, and an F1-score of 0.92 for PD vs. healthy control classification, while 67.15% balanced accuracy was obtained for the more challenging PD vs. differential diagnosis task. Feature analysis showed that PD vs. healthy control discrimination is dominated by tremor-related spectral and amplitude features, whereas variability-related features are more relevant for differential diagnosis. These results demonstrate that competitive performance can be achieved using a simple and interpretable pipeline, providing a practical alternative to more complex deep learning approaches.

Vehicular Ad-Hoc Networks (VANETs) are vital for intelligent transport systems, yet they are extremely vulnerable to cyberattacks like Sybil, Blackhole, Denial-of-Service, Position Falsification, Message Suppression, and Replay. Existing intrusion detection systems use rigid metrics or constrained machine learning, resulting in unreliable detection across different threats. Critically, no previous work combines Zero-Trust security with dynamic traffic management in one vehicle framework. We introduce DATZ-TM (Dynamic Adaptive Zero-Trust Traffic Management), which consists of four integrated components: a calibrated behavioural Trust Scorer with temporal decay, a behavioural Anomaly Detector that couples a learned detection model with a Temporal Graph Neural Network (T-GNN), a Zero-Trust Enforcement layer, and a DQN + LLM traffic agent (Llama-3.3-70B). We evaluate the framework against ten baselines, spanning deep graph (GCN, GAT), temporal (LSTM, TCN), graph-anomaly (Graph-AE) and CNN-LSTM detectors, in two complementary settings: a controlled synthetic benchmark and real SUMO-collected traces of the Luxembourg road network (261 vehicles, 133,773 records), both under an observable-feature-only protocol with per-baseline threshold calibration for fairness and all results across three random seeds. On the synthetic benchmark DATZ-TM attains the best overall performance, leading every baseline on detection rate (86.0%), F1-score (88.5%), AUC (0.955), Matthews Correlation Coefficient (0.788), balanced accuracy (89.2%) and G-mean (89.1%) at a low 7.7% false-positive rate, surpassing the strongest baseline (Random Forest: F1 87.3%) on every aggregate metric. On the harder real-SUMO traces DATZ-TM remains the top detector on the detection-oriented metrics (F1 70.6% vs. Random Forest 69.1%; detection rate 60.7% vs. 56.9%; best balanced accuracy and G-mean) and is competitive on the precision-oriented metrics, while the deep graph baselines degrade sharply (GCN/GAT F1 51/42%). Beyond i.i.d. detection, the temporal model is necessary for coordinated temporal attacks (pulsing/on-off, regime-shift, low-and-slow), where it raises detection by +56.9 percentage points over a no-neural variant (e.g. pulsing attacks 21.6% $$\rightarrow$$ 100%) at a matched 5% false-positive rate. On real SUMO/TraCI simulation the DQN controller outperforms fixed-time and actuated control and is competitive with a max-pressure reference, and the LLM audit layer produces reliable, format-compliant records at $${\approx }0.4$$ s latency.

Man-in-the-Middle (MITM) attacks are common in insecure networks where an attacker can eavesdrop or intercept or modify traffic between two devices. In this work, we performed a practical simulation of a MITM attack using a virtual machine setup. Kali Linux was used as the attacker machine (host/hacking machine) and a Windows VM was used as the target (target machine). First, the attack was executed on a normal, open virtual network, where packet capture was successful. Later, two defense techniques were tested: VPN encryption on the target machine and network segmentation. After enabling these defenses, the attacker was unable to capture or read the traffic. The main aim of this study is to show, in a simple and practical way, how MITM attacks work and how VPNs and network segmentation can stop them. This paper can also be used as an instructional guide for those who want to understand MITM attacks and basic defensive mechanisms in a virtual environment.

Industrial digitalization in emerging economies is frequently constrained by the high cost and rigidity of proprietary Industry 4.0 solutions. This work proposes and experimentally evaluates a low-cost edge gateway architecture that integrates an entry-level PLC, a Raspberry Pi, MQTT messaging and a cloud database to enable basic Industrial Internet of Things (IIoT) functions in Salvadoran manufacturing plants. The architecture is validated in a packaging line case study, where production counting data are acquired from a Siemens LOGO! 8 PLC, processed on a Raspberry Pi using Node-Red and Python, and published to a Supabase cloud backend for visualization and analysis. Over a set of controlled experiments, the prototype achieves a counting error below 1.2%, a 95th-percentile end-to-end latency below 250ms and data delivery above 99.5%, while remaining compatible with the RAMI 4.0 reference model and common OT/IT separation practices. The results indicate that the proposed approach can serve as a practical and affordable first step towards IIoT interoperability in Salvadoran industry, addressing a critical gap in resource-constrained environments.

Ultrasound tomography is a widely adopted diagnostic imaging modality that has undergone rapid advancements in detecting sub-wavelength tumors without the need for ionizing radiation, making it a preferable alternative to the current gold standard of X-ray mammography. This imaging technique relies on an inverse scattering algorithm that utilizes material properties, including sound contrast or attenuation, to detect and identify small targets. In this work, we propose the use of the distorted Born iterative (DBI) method, which is an efficient diffraction tomography approach based on the first-order Born approximation, to improve the biomedical signal processing techniques in ultrasound tomography. Specifically, we investigate the efficacy of various signal processing techniques, including interpolation, dual-frequency, multi-resolution, and multi-frequency DBI (MR-MF-DBI), through numerical simulations. This study serves as a proof-of-concept and will be further expanded using experimental data before clinical application.

Introduction: Nowadays, Shear Wave Elastography (SWE) images are extensively employed for the early detection of various cancers including liver, breast, thyroid, and prostate cancers. This plays a pivotal role in cancer management. SWE images rely on estimating the Complex Shear Modulus (CSM), which is contingent upon two parameters of tissue properties in vivo: elasticity and viscosity. Problem: Reflected noise, random noise, and speckle noise affect the quality of ultrasound images. Objective: Simulating the shear wave propagation by using Finite-Difference Time-Domain (FDTD) method, removing noises, and reconstructing the elasticity and viscosity images of tissues. Methodology: In this study, we utilized several noise filters to eradicate unwanted noise. We employed a directional filter to eliminate reflected noise, an LMS filter to remove random noise, and a median filter for speckle noise removal. Subsequently, we applied the AHI algorithm to estimate the elasticity and viscosity of tissue. Results: The elasticity and viscosity 2D images are estimated, underwent filtering resulting in improved images quality. Root Mean Square Error (RMSE) was used to evaluate the effectiveness of the estimation. Conclusion: The obtained results include 2D images of tissue elasticity and viscosity, demonstrating the effectiveness of the filter in noise removal and AHI algorithm in estimating. Originality: We propose filters for noise removal (consist of reflected noise), before applying AHI algorithm to estimate elasticity and viscosity of tissue. Limitation: The model is applied with a limited range of parameter values in the input scenario.

High transaction fees, security vulnerabilities, and inefficiencies in traditional blockchain systems have long hindered the scalability and cost-effectiveness of real-time sensor networks in the Smart buildings. This study addresses these critical issues by introducing a novel integration of sensor nodes with IOTA-based Distributed Ledger Technology (DLT) networks. The proposed system not only reduces operational costs and network overhead but also enhances security through secure device authentication and communication. Employing state-of-the-art encryption techniques ensures secure and confidential data transmission between sensor nodes and the DLT network with cryptographic operations (0.00017s encryption, 0.00016s decryption), creating a tamper-resistant exchange system. Framed within the context of smart buildings, this research explores scalable and interconnected solutions designed to meet the dynamic security and performance needs of urban environments. The experimental validation demonstrates the system's effectiveness through comprehensive analysis of encryption and decryption performance, network transaction throughput (0.331 TPS), low latency (0.0168s), and energy-efficient power consumption across two distinct scenarios: (1) Normal Transaction conditions and (2) Traffic Congestion environment, where the system shows resilience despite 68% throughput reduction and significant latency increases under network stress.

The Internet of Things (IoT) is experiencing rapid growth, with projections indicating that the number of IoT devices will surpass 40.6 billion by 2034. This expansion, however, is accompanied by the potential threat posed by advancements in quantum computing. The security protocols of IoT devices must undergo significant modifications to effectively address emerging threats. Currently, IoT systems predominantly employ RSA and ECC. However, these cryptographic methods are vulnerable to quantum attacks facilitated by Shor’s algorithm. Centralized IoT systems are vulnerable to single points of failure and data integrity issues. This study proposes a security framework that integrates SPHINCS+ digital signatures, which are known for their resistance to quantum attacks, with blockchain technology for device registration and verification. The system can process approximately 19 transactions per second with an average latency of 94.7 ms. The execution speeds of smart contract operations varied, with the addTransaction operation demonstrating the greatest consistency, averaging 49.7 ms. The deployContract operation requires 69.8 ms because of its complexity. The SPHINCS+ signature operations have an average duration of 42.2 ms. However, this delay can extend to 52-53 ms under conditions of high activity. The framework demonstrates that post-quantum cryptography can be integrated into blockchain-based IoT systems with minimal impact on their performance. This integration provides a practical solution for protecting future IoT systems from both traditional and quantum security threats.

Monitoring and classifying cow behaviors provides valuable support for livestock management. This can be done through sensors attached to the pet. Due to their small size, light weight, and high accuracy, accelerometers are well-suited for this purpose. However, the complexity of behaviors, which often involve similar movements, poses challenges in interpreting the sensor data. This paper presents a novel classifier design for cow behaviors based on acceleration data and a specific set of features. By analyzing cow acceleration data, we extracted features for classification with the help of machine learning algorithms. With five features—Mean, Standard Deviation, Root Mean Square, Median, and Range—and a 15-second data window (1 sample/second), the classifier achieved optimal performance when identifying six behaviors: Feeding, Lying, Standing, Lying-standing-transition, Normal-walking, and Active-walking. The results were validated with public acceleration data. The performance of the proposed classifier has been compared with existing models to highlight the research advantages.

Mood is a prevailing emotional state that remains consistent over time and throughout many settings, reflecting an individual’s overall well-being and outlook. Precisely identifying mood is essential for enhancing human-machine interaction, since it allows for tailored recommendations and a deeper understanding of psychology. Artificial intelligence greatly enhances the accurate detection of mood by evaluating facial expressions and other physiological indicators. Although AI can accurately identify mood, it is as important to control and handle these emotions. Music possesses a formidable capability to manipulate and regulate emotions once it is acknowledged. This study presents a streamlined AI-powered approach for identifying and influencing moods by leveraging music. Furthermore, a camera linked to Internet of Things (IoT) device is incorporated into the cloud-based framework to recognize emotions and provide responses in the form of music. Ecosystem evaluation is performed by utilizing numerous factors.

In the recruitment process, manually selecting suitable candidates from curriculum vitae (CVs) for a job description (JD) is both time-consuming and expensive. Traditional keyword-based methods struggle to capture skill semantics, prompting the development of more advanced JD-CV matching systems. This paper aims to investigate and construct an ontology-based skills recommendation system, with objectives including creating a skills ontology and developing skills matching methods for JD-CV pairs. The objective of our approach is to enhance the accuracy and contextual relevance of recommendations by utilizing the proposed score. The proposed skills ontology and skills matching strategies are applied to a real dataset in Vietnam. The results of our study can automatically recommend a list of CVs for a given JD. Furthermore, the findings indicate that our proposed model surpasses comparative approaches by a margin of at least 1% to 5%. Overall, the study demonstrates the potential of utilizing ontology-based approaches to offer a practical solution for enhancing hiring practices.