Journal of Computers, Mechanical and Management

Artificial Intelligence and Machine Learning in Precision Medicine

2025-03-26T07:07:02+03:00

Artificial intelligence (AI) and machine learning (ML) are transforming healthcare delivery by facilitating the development of precision medicine, which prioritizes personalized diagnostic and treatment strategies based on individual genetic, physiological, and lifestyle profiles. This study investigates the contributions of AI and ML in enhancing clinical decision-making, improving diagnostic accuracy, and supporting remote patient management. A mixed-methods framework was applied, combining quantitative analysis of clinical datasets with qualitative interviews and real-world case evaluations. Machine learning algorithms, including convolutional neural networks and ensemble models, were trained on public datasets to assess their impact on diabetes and cardiovascular care. Results showed significant improvements in glycemic control and reductions in hospital readmissions, indicating effective treatment personalization. Semi-structured interviews with patients and healthcare professionals revealed strong support for AI-enabled tools, highlighting perceived benefits such as increased efficiency, ease of use, and diagnostic clarity. Case studies of wearable health devices and telemedicine systems demonstrated enhanced care accessibility and a reduction in in-person clinical consultations. Ethical and operational challenges were identified as key concerns. Issues such as data privacy, algorithmic bias, lack of explainability, and the need for sustained human oversight were recurrent themes in stakeholder feedback. These challenges underscore the necessity of implementing transparent, accountable, and ethically grounded AI systems in clinical practice. The study underscores the dual necessity of technological capability and ethical rigor in deploying AI for precision medicine. Through a comprehensive analysis of clinical, experiential, and operational data, the research highlights both the promise and the complexity of integrating AI in modern healthcare environments.

AI-Driven Decision Support System for Multidimensional Academic Performance Prediction in Higher Education

2025-03-25T16:29:04+03:00

The increasing integration of artificial intelligence (AI) into educational systems has highlighted the limitations of traditional data analysis tools in academic performance assessment. This study proposes a four-level AI-enhanced Decision Support System (DSS) employing Artificial Neural Networks (ANN) to classify and predict student outcomes based on multi-semester academic data and co-curricular attributes. The dataset, comprising information from 300 students, includes academic scores, participation in extracurricular activities, and skill assessments. Data preprocessing and feature selection strategies were implemented to optimize model input. The ANN model achieved high accuracy across three semesters, providing granular and actionable insights for educators. The system further identifies individual and cohort-level trends, supports personalized feedback, and enables proactive intervention strategies. The proposed DSS demonstrates a scalable, interpretable, and effective approach for performance analysis in contemporary educational settings.

Advanced Water Resource Management Using IoT and Big Data Analytics

2025-03-25T13:19:02+03:00

Effective water resource management is increasingly essential in mitigating the impacts of water scarcity and environmental degradation. This study proposes an integrated system that leverages the Internet of Things (IoT) and Big Data Analytics to enhance efficiency, responsiveness, and sustainability in water governance. The methodology includes real-time data collection through smart sensors, application of statistical and machine learning techniques for predictive modeling, and blockchain-backed data management for transparency. A 30-day simulation involving 50 sensor nodes demonstrated improvements including a 20% enhancement in water quality and a 7% reduction in daily usage. The outcomes validate the viability of this approach, aligning with sustainable development goals and supporting intelligent decision-making in both urban and agricultural contexts.

Simulation-Guided Synthesis and Evaluation of Advanced Nanomaterials for Environmental Remediation

2025-03-24T12:39:39+03:00

This study presents a simulation-guided strategy for the synthesis, characterization, and environmental application of advanced nanomaterials, aiming to address the growing concerns of pollutant accumulation in air, water, and soil matrices. The research leverages atomistic and electronic modeling tools, including Molecular Dynamics (MD) and Density Functional Theory (DFT), to identify and optimize structural and thermodynamic parameters critical for nanomaterial efficacy. Simulations performed using platforms such as LAMMPS, GROMACS, VASP, and Quantum ESPRESSO were instrumental in predicting nanoparticle stability, surface energy, and reactivity under environmentally relevant conditions. The study further incorporates environmental transport modeling via COMSOL Multiphysics to predict contaminant flow and interaction with the synthesized nanostructures. Experimentally, nanomaterials synthesized through hydrothermal, sol-gel, and chemical precipitation routes were characterized using SEM, XRD, and FTIR. Surface area and morphology analyses revealed that the nanostructures possessed high porosity and uniform distribution with an average particle size of 30 nm and a specific surface area of 250 m²/g. The adsorption studies showed pollutant removal efficiencies of 95% for heavy metals and 90% for organic compounds, with an adsorption capacity of 500 mg/g. These performance metrics are indicative of favorable kinetics, supported by pseudo-second-order models suggesting chemisorption as the dominant removal mechanism. The findings demonstrate that simulation-informed synthesis can systematically guide material development toward achieving optimal interaction with environmental pollutants. The combined use of in silico and experimental approaches ensures both predictive robustness and empirical validation. This hybrid framework not only enhances the functional reliability of nanomaterials but also accelerates the development of environmentally sustainable technologies. The approach presented herein offers a scalable path toward the deployment of nanotechnology in large-scale remediation operations, contributing meaningfully to pollution control and ecosystem restoration.

Blockchain-Based Decentralized Storage for Scalable and Secure IoT Data Management

2025-03-26T15:49:52+03:00

The rapid expansion of Internet of Things (IoT) ecosystems has resulted in an unprecedented surge in data generation, necessitating reliable, scalable, and secure storage mechanisms. Traditional centralized storage systems suffer from inherent limitations such as single points of failure, limited scalability, and vulnerability to cyberattacks, which compromise the confidentiality and availability of critical IoT data. This study introduces a blockchain- based decentralized storage framework aimed at addressing these critical issues. By leveraging the distributed and immutable characteristics of blockchain technology, the proposed system enhances data integrity, ensures transparency, and facilitates trustless data exchange among heterogeneous IoT devices. The methodology includes mathematical modeling of key performance parameters such as latency, throughput, storage efficiency, and consensus delay. Smart contracts are integrated to automate validation and enforce rules among interconnected devices, while redundancy mechanisms like replication and erasure coding improve storage reliability and efficiency. The framework’s effectiveness is evaluated using simulation tools including Hyperledger Caliper and Ethereum Testnets for blockchain behavior, and NS-3 and OMNeT++ for modeling dynamic IoT network environments. Experimental results reveal a 30% improvement in data retrieval time, 25% gain in storage efficiency, 40% enhancement in system resilience, and a 50% increase in transaction throughput over conventional approaches. These metrics highlight the suitability of the proposed model for real-world applications requiring scalable and secure IoT data management, such as healthcare monitoring, smart cities, and industrial automation. The model’s reproducibility and modularity make it a robust solution for future research and deployment. Overall, this work demonstrates that blockchain-integrated decentralized storage frameworks present a transformative step toward resilient and scalable IoT infrastructures.