Abubakar Elsafi^1,*, Arafat Abdulgader Mohammed Elhag², Lubna A. Gabralla³, Ali Ahmed⁴, Ashraf Osman Ibrahim⁵

CMES-Computer Modeling in Engineering & Sciences, Vol.145, No.1, pp. 621-647, 2025, DOI:10.32604/cmes.2025.072726 - 30 October 2025

Abstract Energy efficiency stands as an essential factor when implementing deep reinforcement learning (DRL) policies for robotic control systems. Standard algorithms, including Deep Deterministic Policy Gradient (DDPG), primarily optimize task rewards but at the cost of excessively high energy consumption, making them impractical for real-world robotic systems. To address this limitation, we propose Physics-Informed DDPG (PI-DDPG), which integrates physics-based energy penalties to develop energy-efficient yet high-performing control policies. The proposed method introduces adaptive physics-informed constraints through a dynamic weighting factor (), enabling policies that balance reward maximization with energy savings. Our motivation is to overcome the… More >

Hui Qi¹, Jingyi Lian², Congjun Rao^2,*

CMES-Computer Modeling in Engineering & Sciences, Vol.145, No.1, pp. 995-1028, 2025, DOI:10.32604/cmes.2025.069722 - 30 October 2025

Abstract For the past few years, the prevalence of cardiovascular disease has been showing a year-on-year increase, with a death rate of 2/5. Coronary heart disease (CHD) rates have increased 41% since 1990, which is the number one disease endangering human health in the world today. The risk indicators of CHD are complicated, so selecting effective methods to screen the risk characteristics can make the risk prediction more efficient. In this paper, we present a comprehensive analysis of CHD risk indicators from both data and algorithmic levels, propose a method for CHD risk indicator identification based… More >

Md Sabir Hossain¹, Md Mahfuzur Rahman^1,2,*, Mufti Mahmud^1,3

CMES-Computer Modeling in Engineering & Sciences, Vol.145, No.1, pp. 1087-1116, 2025, DOI:10.32604/cmes.2025.068779 - 30 October 2025

Abstract This article presents a human fall detection system that addresses two critical challenges: privacy preservation and detection accuracy. We propose a comprehensive framework that integrates state-of-the-art machine learning models, multimodal data fusion, federated learning (FL), and Karush-Kuhn-Tucker (KKT)-based resource optimization. The system fuses data from wearable sensors and cameras using Gramian Angular Field (GAF) encoding to capture rich spatial-temporal features. To protect sensitive data, we adopt a privacy-preserving FL setup, where model training occurs locally on client devices without transferring raw data. A custom convolutional neural network (CNN) is designed to extract robust features from More > Graphic Abstract

Si-Yu Gao¹, Siu-Man Ng^2,*

International Journal of Mental Health Promotion, Vol.27, No.10, pp. 1471-1482, 2025, DOI:10.32604/ijmhp.2025.068702 - 31 October 2025

Abstract Background: Existing Chinese stigma scales focus on the perceptions of people with mental illness (PMI) without assessing the general public’s attitudes toward integrating PMI into the community. Developing a valid and reliable Chinese instrument measuring the attitude domain will be helpful to future research in this area. The current study aimed to validate a shortened Chinese version of the Community Attitudes towards the Mentally Ill Scale (C-CAMI-SF). Methods: Four hundred participants who are (1) Chinese; (2) aged 18 years and above; and (3) able to complete the Chinese questionnaire in a self-reported manner participated in… More >

Dan Wang¹, Zhoubin Li¹, Yuze Xia^1,2,*, Zhenhua Yu^1,*

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 5823-5845, 2025, DOI:10.32604/cmc.2025.071656 - 23 October 2025

Abstract Multimodal Aspect-Based Sentiment Analysis (MABSA) aims to detect sentiment polarity toward specific aspects by leveraging both textual and visual inputs. However, existing models suffer from weak aspect-image alignment, modality imbalance dominated by textual signals, and limited reasoning for implicit or ambiguous sentiments requiring external knowledge. To address these issues, we propose a unified framework named Gated-Linear Aspect-Aware Multimodal Sentiment Network (GLAMSNet). First of all, an input encoding module is employed to construct modality-specific and aspect-aware representations. Subsequently, we introduce an image–aspect correlation matching module to provide hierarchical supervision for visual-textual alignment. Building upon these components, More >

Khadija Slimani^1,2,*, Samira Khoulji², Hamed Taherdoost^3,4, Mohamed Larbi Kerkeb⁵

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 4667-4686, 2025, DOI:10.32604/cmc.2025.070173 - 23 October 2025

Abstract The integration of the dynamic adaptive routing (DAR) algorithm in unmanned aerial vehicle (UAV) networks offers a significant advancement in addressing the challenges posed by next-generation communication systems like 6G. DAR’s innovative framework incorporates real-time path adjustments, energy-aware routing, and predictive models, optimizing reliability, latency, and energy efficiency in UAV operations. This study demonstrated DAR’s superior performance in dynamic, large-scale environments, proving its adaptability and scalability for real-time applications. As 6G networks evolve, challenges such as bandwidth demands, global spectrum management, security vulnerabilities, and financial feasibility become prominent. DAR aligns with these demands by offering More >

Boyan Cui^1,2, Huijuan Wang^1,*, Yongjun Qi^1,*, Hongce Chen¹, Quanbo Yuan^1,3, Dongran Liu¹, Xuehua Zhou¹

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 5531-5553, 2025, DOI:10.32604/cmc.2025.070101 - 23 October 2025

Abstract The proliferation of malware and the emergence of adversarial samples pose severe threats to global cybersecurity, demanding robust detection mechanisms. Traditional malware detection methods suffer from limited feature extraction capabilities, while existing Vision Transformer (ViT)-based approaches face high computational complexity due to global self-attention, hindering their efficiency in handling large-scale image data. To address these issues, this paper proposes a novel hybrid enhanced Vision Transformer architecture, HERL-ViT, tailored for malware detection. The detection framework involves five phases: malware image visualization, image segmentation with patch embedding, regional-local attention-based feature extraction, enhanced feature transformation, and classification. Methodologically,… More >

Khawaja Tahir Mehmood^1,*, Muhammad Majid Hussain²

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 5643-5670, 2025, DOI:10.32604/cmc.2025.068404 - 23 October 2025

Abstract In the modern era of 5th generation (5G) networks, the data generated by User Equipments (UE) has increased significantly, with data file sizes varying from modest sensor logs to enormous multimedia files. In modern telecommunications networks, the need for high-end security and efficient management of these large data files is a great challenge for network designers. The proposed model provides the efficient real-time virtual data storage of UE data files (light and heavy) using an object storage system MinIO having inbuilt Software Development Kits (SDKs) that are compatible with Amazon (S3) Application Program Interface (API)… More >

Hanyong Lee^#, Ye-Chan Park^#, Jaesung Lee^*

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 4977-4993, 2025, DOI:10.32604/cmc.2025.068156 - 23 October 2025

Abstract Restoring texts corrupted by visually perturbed homoglyph characters presents significant challenges to conventional Natural Language Processing (NLP) systems, primarily due to ambiguities arising from characters that appear visually similar yet differ semantically. Traditional text restoration methods struggle with these homoglyph perturbations due to limitations such as a lack of contextual understanding and difficulty in handling cases where one character maps to multiple candidates. To address these issues, we propose an Optical Character Recognition (OCR)-assisted masked Bidirectional Encoder Representations from Transformers (BERT) model specifically designed for homoglyph-perturbed text restoration. Our method integrates OCR preprocessing with a… More >

Kunal Thakur¹, Ashu Taneja^1,*, Ali Alqahtani², Nayef Alqahtani³

CMC-Computers, Materials & Continua, Vol.85, No.3, pp. 4827-4848, 2025, DOI:10.32604/cmc.2025.067901 - 23 October 2025

Abstract Accident detection plays a critical role in improving traffic safety by enabling timely emergency response and reducing the impact of road incidents. The main challenge lies in achieving real-time, reliable and highly accurate detection across diverse Internet-of-vehicles (IoV) environments. To overcome this challenge, this paper leverages deep learning to automatically learn patterns from visual data to detect accidents with high accuracy. A visual classification model based on the ResNet-50 architecture is presented for distinguishing between accident and non-accident images. The model is trained and tested on a labeled dataset and achieves an overall accuracy of… More >