Raffles University Iskandar

Purpose The increasing worldwide focus on environmental, social and governance (ESG) principles has compelled higher education institutions (HEIs) to re-evaluate their strategic responsibilities in promoting sustainable development. In Malaysia, the shift from ESG policy frameworks to institutional implementation is uneven and fragmented. Design/methodology/approach This systematic literature review (SLR) consolidates 26 peer-reviewed studies published from 2020 to 2025 to critically analyse the integration of ESG principles by Malaysian higher education institutions across five thematic domains – policy and institutional commitment, curriculum integration, campus operations, stakeholder engagement and governance and reporting. Findings Results indicate that while institutional knowledge of ESG has grown, its implementation often suffers from inconsistency, insufficient theoretical foundation and a lack of national standards. Principal issues include restricted curricular integration, variable environmental practices, superficial stakeholder engagement and disjointed reporting systems. The assessment reveals considerable demographic deficiencies, notably the underrepresentation of private and rural colleges and underscores the lack of customized ESG guidelines for the Malaysian higher education sector. Practical implications This paper advocates for strategic, theory-based and contextually adaptable ESG integration to facilitate Malaysia’s comprehensive sustainable transition. By integrating national educational reforms with ESG principles, higher education institutions may act as catalysts for systemic and inclusive sustainability progress. This report provides practical insights for politicians, educators and institutional leaders aiming to close the gap between policy and practice in ESG adoption. Originality/value This study is the first SLR to critically examine the integration of ESG principles within Malaysia’s higher education sector through a structured thematic synthesis of 26 peer-reviewed publications from 2020 to 2025. Unlike previous works that focus narrowly on sustainability education or isolated campus initiatives, this review holistically maps ESG practices across five core domains – policy, curriculum, operations, stakeholder engagement and reporting – while uncovering theoretical, practical and population-level gaps. By focusing specifically on the Malaysian context, the study provides region-specific insights that are often underrepresented in global ESG discourse. The originality of this paper lies in its framework-driven synthesis and cross-theme evaluation, which collectively offer actionable recommendations for institutional leaders, policymakers and academics aiming to bridge the policy–practice divide in ESG adoption. The value of this work extends beyond scholarly contribution; it provides a strategic lens to support Malaysia’s higher education institutions in aligning with national sustainability goals and global benchmarks such as the United Nations Sustainable Development Goals.

Increased specificity in brain tumor detection significantly improves diagnosis and treatment planning, contributing to improved medical imaging and patient care This article proposes a cluster model for brain tumor detection and uses the Segment Anything (SAM) and You Only Look Once (YOLOv5). It takes advantage of the complementary strengths of both models to give a detection system and is overall well-tested by the Ensemble model on benchmark datasets, such as a complete brain tumor dataset, and has better performance in terms of accuracy, precision, recall, and F1 scores than single models. The results reflect the strength of ensemble models in medical imaging that it can significantly enhance diagnostic techniques and enhance the efficacy of smart healthcare systems.

Deep neural networks (DNNs) achieve remarkable performance in machine learning, but are vulnerable to adversarial attacks, such as those generated by Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD), which induce misclassifications, endangering safety-critical applications like autonomous driving, medical diagnostics, and cybersecurity. This study proposes a two-phase training framework to enhance adversarial robustness without sacrificing accuracy. The initial phase establishes baseline performance, while the adaptive phase iteratively evolves the dataset by incorporating adversarial synthetic data generated via FGSM and PGD, targeting model vulnerabilities through cycles of training, evaluation, and dataset augmentation. Evaluated on the Modified National Institute of Standards and Technology database (MNIST, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 0, 0 0 0}$</tex> samples), the Canadian Institute For Advanced Research 10-class dataset (CIFAR-10, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 0, 0 0 0}$</tex> samples), BrainTumor-7K (7,023 MRIs), and the Phishing Legitimate Dataset (Phishing, <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 0, 0 0 0}$</tex> samples), this framework yields significant robustness gains against FGSM and PGD attacks: 73.73 % on MNIST <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(92.71 \pm 0.32 \%$</tex> vs. <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$18.98 \pm 0.45 \%$</tex> adversarial accuracy), 7.17 % on CIFAR-10 (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$41.04 \pm 0.37 \%$</tex> vs. <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$33.87 \pm 0.41 \%$</tex>), 73.22 % on BrainTumor-7K (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$89.60 \pm 0.33 \%$</tex> vs. <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$16.38 \pm 0.42 \%$</tex>), and 20.85 % on Phishing (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$94.10 \pm 0.22 \%$</tex> vs. <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$73.25 \pm 0.38 \%$</tex>). Furthermore, clean accuracy improves (e.g., 0.20 % MNIST, 3.12 % CIFAR-10), challenging the robustness-accuracy trade-off. This framework provides a scalable and efficient solution for resilient AI, preserving innovative methods for future intellectual property protection in cybersecurity and medical imaging applications

Polycystic Ovary Syndrome (PCOS) is an important endocrine and metabolic disorder that affects women in their reproductive years. It is characterized by the increase in male hormones, no ovulation, and the formation of many cysts in the ovaries. Standard treatment consists of using oral contraceptives, insulin sensitizers and anti-androgens, among others, which only manage the symptoms and not the underlying causes, besides, they usually have negative side effects and their long-term efficacy is often limited. Moreover, they usually come up with side effects and have limited effectiveness in the long run. On the other hand, the work of bioactive phytochemicals, including flavonoids, polyphenols, alkaloids, terpenoids, and saponins, has provided the hope in the area of treating the disease because they can restore the endocrine system's control, lower the oxidative stress, and enhance the insulin sensitivity. The recent developments in nanocarrier systems, such as polymeric nanoparticles and antibody-conjugated formulations, are having a beneficial impact on the stability, bioavailability, and targeted delivery of the phytochemicals. Furthermore, the understanding of modulation of gut microbiota, functional genomics, and proteomic interactions has paved the way for new therapeutic pathways in the management of PCOS. This chapter presents a discussion on the technical challenges in the usage of nanotechnology and phytochemistry, the translational challenges, and the future of PCOS therapies that are safe, effective, and precision-based.

The United Arab Emirates (UAE) has undergone a remarkable transformation, evolving from modest beginnings into a global business and tourism hub, with sustainability at the core of its development strategy. As the UAE’s Emirates, particularly Abu Dhabi and Dubai, have grown rapidly, the government has recognized the need for customized urban sustainability guidelines to address the unique environmental challenges of the region. The introduction of the Green Building and Sustainable Building code is a testament to this commitment, targeting significant reductions in carbon emissions and promoting resource efficiency across the country’s infrastructure. In the sports sector, the UAE’s modern facilities and expansive infrastructure reflect a strong dedication to sustainability. The government has implemented ambitious green targets, encouraging sports venues and events to adopt energy-efficient technologies and environmentally friendly practices. The Department of Culture and Tourism (DCT) has been instrumental in this effort, introducing sustainable procurement policies that incentivize event organizers to prioritize green initiatives. High-profile events like Expo 2020 Dubai and COP28 have highlighted the UAE’s leadership in sustainability, showcasing innovative approaches to environmental stewardship. Despite the challenges of balancing rapid economic growth with sustainable development, the UAE’s approach offers valuable insights for other nations, particularly in the role of strong leadership and strategic planning in driving sustainable practices. As the UAE continues to embed sustainability into its policies and practices, it sets a powerful example of how a nation can future-proof its growth while fostering a culture of environmental responsibility.

Maintaining consistency in automated essay scoring is essential to guarantee fair and dependable assessments. This study investigates consistency and provides a comparative analysis of open-source and proprietary large language models (LLMs) for automated essay scoring (AES). The study utilized student essays, each assessed five times to measure both intrarater (using intraclass coefficient and repeatability coefficient) and interrater (concordance correlation coefficient) reliability across several models: GPT-4, GPT-4o, GPT-4o mini, GPT-3.5 Turbo, Gemini 1.5 Flash, and LLaMa 3.1 70B. Essays and marking criteria are used for prompt construction and sent to each large language model to obtain score outputs. Results indicate that the scores generated by GPT-4o closely align with human assessments, demonstrating fair agreement across repeated measures. Specifically, GPT-4o exhibits slightly higher concordance correlation coefficients (CCC) than GPT-4o mini, indicating superior agreement with human scores. However, qualitatively, it can be observed that all LLM models are not as consistent in terms of their scoring rationale/evaluation. Our study results indicate that the challenges currently faced in automated essay scoring with large language models need to be analyzed not only from a quantitative perspective but also qualitatively. Additionally, we utilize more sophisticated prompting methods and address the inconsistencies observed in the initial measurements. Despite the purported reliability of some models within our study, the selection of LLMs should be considered thoroughly during practical implementations for an AES.