University of Asia Pacific

Accurate and early detection of brain tumors is essential for effective treatment planning in medical diagnosis. However, deep learning (DL) models often struggle with MRI-based tumor detection due to significant variability in tumor size, shape, and location. Traditional diagnostic techniques are limited by subjectivity and low interpretability, while many DL models operate as black boxes, reducing clinical trust. Incorporating attention mechanisms can help by directing the model's focus to the most informative regions of an image, thus improving both accuracy and interpretability. However, existing attention methods often fail to capture the complex spatial and contextual features present in medical images such as MRI scans. In this study, we propose a novel attention-based, explainable DL framework designed to improve the performance and transparency of brain tumor diagnosis. We introduce the Strip-Style Pooling Attention Network (SSPANet), which combines the strengths of channel and spatial attention mechanisms to more effectively capture intricate imaging features. We evaluated SSPANet using VGG16 and ResNet50 as backbone architectures, integrating it alongside existing attention methods for comparison. Among all configurations, ResNet50 combined with SSPANet achieves the best results, with 97% accuracy, precision, recall, and F1-score, along with 95% Cohen's Kappa and Matthews Correlation Coefficient. For interpretability, we employ GradCAM, GradCAM++, and EigenGradCAM across attention-guided DL models. The ResNet50 + SSPANet + GradCAM++ combination consistently provides superior visual explanations, highlighting SSPANet's ability to capture complex spatial-contextual information effectively. We also offer a theoretical analysis to support the efficiency and effectiveness of the proposed attention mechanism.

Fueled by the help of the Internet, more and more tourists nowadays tend to plan trips by themselves in order to have trip plans that meet their preferences and convenience perfectly. However, tourists may face some problems when they plan a trip by themselves, which makes the whole trip planning process challenging and tiring. These problems include extensive tourism information, manually constructing an itinerary for a trip and difficulty in satisfying the needs of all trip participants. Therefore, a web-based trip planning system is proposed in this project to solve all these problems in order to help tourists in planning their desired trips more effectively and efficiently. This system will help users to search for attractions and restaurants in Southeast Asian countries faster by providing filtering and prioritizing features. This system also facilitates the decision-making process of tourists when choosing places to visit and restaurants by utilizing the power of word-of-mouth (reviews). Besides, this system will aid tourists by reducing the need to manually construct the itinerary for a trip. The k-means clustering algorithm will be used to auto-arrange the trip itinerary to ensure places close to each other are arranged to be visited on the same day so that tourists can save on unnecessary transport costs and time. Lastly, this system promotes collaborative trip planning by providing a platform for all the participants in a trip to discuss and plan their trip together.

Emerging brain-computer interface (BCI) technologies powered by artificial intelligence (AI) can retrieve, store, and analyze human neuro data. Originally designed for managing disabilities disconnecting the brain from motor neural control, these technologies are being developed for wider use without clear ethical frameworks. This paper examines ethical and legal implications of AI-based BCIs and evaluates their prospects in Japan. It presents findings from a national survey of 2000 Japanese citizens regarding healthcare AI perceptions and another survey of 228 university students from Japan and other Asian countries regarding neuro-ethics and neuro-privacy awareness. The national survey includes 36 questions about citizens’ healthcare needs, services, AI familiarity, AI use in healthcare, new technologies, and ethical issues surrounding healthcare AI. Analysis demonstrates the Japanese population has limited AI and smart technology knowledge, trusts the healthcare system and physicians to use it properly, and has few concerns about potential pitfalls of new healthcare technologies. Nevertheless, they worry about privacy intrusion and prefer human resources over AI in healthcare. The university survey reveals Japanese youth are less aware of and concerned about ethical risks associated with BCIs compared with peers from other Asian countries. This reveals a notable gap between public perceptions and ethical and legal challenges highlighted in literature, underscoring the need for governance frameworks safeguarding core dimensions of the human condition, including autonomy, dignity, and self-determination. The study recommends AI-assisted medical tools be audited by engineers, ethicists, and legal experts, guided by human-centered ethical values and deployed under physician oversight.

The discovery and ethnomedicinal applications of Lepisanthes rubiginosa were elucidated through comprehensive ethnobotanical observations among indigenous populations residing in forest and semi-forest ecosystems. It was utilized for its key role as medicament for long time with medicinal and economic value. Therefore, the primary objective of the current study was to perform a comprehensive phytochemical analysis, antimicrobial activity, chromatographic analysis and chemical profiling of L. rubiginosa leaf extract. The antibacterial activity was assessed using the disc diffusion method and in chromatographic analysis TLC was utilized to separate the methanolic extract using a solvent system consisting of n-Hexane, Ethyl Acetate (EtOAc), methanol (MeOH) and chloroform (CHCl3). Moreover, methanolic extract was characterized using FTIR-ATR and GC-MS. Phytochemical screening showed the presence of sugar, glycosides, alkaloids, tannins, phenolic content (106.57 f 11.25 mg GAE/g) and flavonoid content (88.72 f 3.34 QE mg/g). In antimicrobial assessments, the dose of 500 mu g/disc showed the highest inhibition against the gram-negative bacterium Shigella boydii and the inhibitory zone measured 26.2 f 1.2 mm whereas in Bacillus subtilis, a gram-positive bacterium, the dosage of 250 mu g/disc showed the least amount of inhibition and that was 9.0 f 0.4 mm. TLC results showed ethyl acetate: n-hexane (4,1) ratio separates five compounds from the extract and their Rf values were 0.14, 0.22, 0.33, 0.48, and 0.82 respectively. FTIR analysis confirms the presence of functional groups like aldehydes, ketones, ethers, esters, amines, alcohols etc. Twelve compounds were identified using GC-MS, with the highest percentage (64.10%) being phenol, 2,4-bis(1,1-dimethylethyl). Further investigation will be needed for isolation and characterization of bioactive compounds.

The assessment of Japan’s security policy during the first 23 years of the twenty-first century went through a 180-degree turn from that of a reactive pacifist to a proactive strategic player. With the end of the Cold War and the emergence of new challenges at a global and regional level, Japan’s leading political drivers, summed up in the Yoshida Doctrine, seemed unable to address the new geopolitical context, entering a phase of reform. The late Prime Minister Abe Shinzo’s unusually strong and long leadership between 2012 and 2020 undoubtedly contributed to this shift. While the role of prime ministers in post-Abe years appears to have reverted to a more traditional facilitator role, the strategic direction Abe set in such initiatives as the Quadrilateral meetings among the United States, Japan, Australia and India, as well as the enhancement of Japan’s participation in collective defence and security, remain largely intact in rhetoric. Exploring how this rhetoric exercise translated into more tangible security policies, the article examines Japan’s security diplomacy and defence strategy in the post-Abe context.