Experimental analysis of the workability of copper matrix composites enhanced with MoS2 and TiO2 particles Mohamed Kchaou, Velayudham Subbian, Sujin Jose Arul, Hariharan Nair Sandeep, Faisal Khaled Aldawood, et al. International Journal of Materials Research, 2026 This study investigated the workability of copper-based hybrid composites. The copper composites with enhanced mechanical properties were fabricated using powder metallurgy to be reinforced with titanium dioxide (TiO 2 ) and molybdenum disulfide (MoS 2 ). Triaxial stress state conditions were used to evaluate the performance of the copper composites, such as true axial strain and stress, true hoop stress, true mean stress, true effective stress, strain hardening index, strength coefficient, and instantaneous strain hardening from the measurement obtained using cold upset testing. The relationships between various stresses and stress–strain ratios were plotted and analyzed. Results showed that the increased reinforcement contents in the copper matrix enhanced the mechanical properties of the copper composites, especially the true axial, the true hoop, and the true effective stresses. Furthermore, the combination of 5 wt.% of TiO 2 and 4 wt.% of MoS 2 in the copper matrix was identified as the optimum composition for the best workability.
Artificial Intelligence and Machine Learning in Tribology: Selected Case Studies and Overall Potential Raj Shah, Rudy Jaramillo, Garvin Thomas, Thohid Rayhan, Nayem Hossain, et al. Advanced Engineering Materials, 2025 Artificial intelligence (AI) and machine learning (ML) have been the subjects of increased interest in recent years due to their benefits across several fields. One sector that can benefit from these tools is the tribology industry, with an emphasis on friction and wear prediction. This industry hopes to train and utilize AI algorithms to classify equipment life status and forecast component failure, mainly using supervised and unsupervised learning approaches. This article examines some of the methods that have been used to accomplish this, such as condition monitoring for predictions in material selection, lubrication performance, and lubricant formulation. Furthermore, AI and ML can support the determination of tribological characteristics of engineering systems, allowing for a better fundamental understanding of friction, wear, and lubrication mechanisms. Moreover, the study also finds that the continued use of AI and ML requires access to findable, accessible, interoperable, and reusable data to ensure the integrity of the prediction tools. The advances of AI and ML methods in tribology show considerable promise, providing more accurate and extensible predictions than traditional approaches.
Industry 5.0 adaptation for disability-inclusive healthcare: A review of emergent and AI technologies for assistive digital health Mohamed Kchaou, Yamuna Munusamy, Khalid Ayed Alharthi, Akram Fadhl Al-mahmodi Digital Health, 2025 Industry 5.0 is reshaping healthcare through human-centric design, sustainability, and advanced technologies. However, there is limited insight into how these innovations address the specific needs of people with disabilities. This review aims to examine the role of emerging and AI-driven technologies in enabling disability-inclusive digital healthcare solutions. A comprehensive scoping review was conducted, focusing on studies published in recent years on Industry 5.0 technologies applied to disability-inclusive digital healthcare pathways. Key technologies reviewed include collaborative robotics, virtual reality, telemedicine, and human-centered artificial intelligence. Relevant case studies and ethical considerations were also analysed. The analysis highlighted that Industry 5.0 technologies show promise in enhancing diagnostic accuracy, personalization, and accessibility for people with disabilities. Applications include remote assessments, assistive tools, and adaptive interfaces that improve diagnostic processes. Despite this progress, integration of these technologies remains fragmented, and challenges such as ethical concerns, regulatory barriers, and inclusive design persist. This review uniquely synthesizes these technologies within the framework of Industry 5.0, offering a broader perspective than prior single-technology reviews and proposing a roadmap for the successful implementation that incorporates training, regulatory alignment, interdisciplinary collaboration, social-economic barriers, real-world evidence, and inclusivity across disability types. As conclusion, Industry 5.0 holds significant promise for advancing disability-inclusive digital healthcare. Realizing this potential, however, requires coordinated efforts to address integration gaps, strengthen ethical and regulatory frameworks, and embed user-centered co-design principles. Future research should focus on more developing inclusive, and sustainable diagnostic solutions aligned with Industry 5.0 principles.
Robust Autism Spectrum Disorder Screening Based on Facial Images (For Disability Diagnosis): A Domain-Adaptive Deep Ensemble Approach Mohammad Shafiul Alam, Muhammad Mahbubur Rashid, Ahmad Jazlan, Md Eshrat E. Alahi, Mohamed Kchaou, et al. Diagnostics, 2025 Background/Objectives: Artificial intelligence (AI) is revolutionising healthcare for people with disabilities, including those with autism spectrum disorder (ASD), in the era of advanced technology. This work explicitly addresses the challenges posed by inconsistent data from various sources by developing and evaluating a robust deep ensemble learning system for the accurate and reliable classification of autism spectrum disorder (ASD) based on facial images. Methods: We created a system that learns from two publicly accessible datasets of ASD images (Kaggle and YTUIA), each with unique demographics and image characteristics. Utilising a weighted ensemble strategy (FPPR), our innovative ASD-UANet ensemble combines the Xception and ResNet50V2 models to maximise model contributions. This methodology underwent extensive testing on a range of groups stratified by age and gender, including a critical assessment of an unseen, real-time dataset (UIFID) to determine how well it generalised to new domains. Results: The performance of the ASD-UANet ensemble was consistently better. It significantly outperformed individual transfer learning models (e.g., Xception alone on T1+T2 yielded an accuracy of 83%), achieving an impressive 96.0% accuracy and an AUC of 0.990 on the combined-domain dataset (T1+T2). Notably, the ASD-UANet ensemble demonstrated strong generalisation on the unseen real-time dataset (T3), achieving 90.6% accuracy and an AUC of 0.930. This demonstrates how well it generalises to new data distributions. Conclusions: Our findings demonstrate significant potential for widespread, equitable, and clinically beneficial ASD screening using this promising, reasonably priced, and non-invasive method. This study establishes the foundation for more precise diagnoses and greater inclusion for people with autism spectrum disorder (ASD) by integrating methods for diverse data and combining deep learning models.
3D-Printed Objects for Multipurpose Applications Nayem Hossain, Mohammad Asaduzzaman Chowdhury, Md. Bengir Ahmed Shuvho, Mohammod Abul Kashem, Mohamed Kchaou Journal of Materials Engineering and Performance, 2021
