Hsing Wu University

Type 2 diabetes (T2D) is associated with cognitive decline and neurodegenerative disorders. Changes in the connections between brain regions responsible for emotions and memory might play a role in the reduced cognitive function observed in individuals with diabetes. Utilizing machine learning approaches on neuroimaging data shows potential for exploring these intricate associations. A research study was carried out with 40 individuals diagnosed with T2D and 30 control participants, all of whom were middle-aged and right-handed. The participants underwent neuropsychological assessments and fMRI scans while engaging in an emotional Stroop task. Our analysis concentrated on the functional connectivity of specific brain regions associated with cognitive control. We utilized a fully connected network (FCN)-based machine learning approach to predict cognitive decline using neural connectivity patterns. The FCN accurately forecasted Montreal Cognitive Assessment scores in patients with Type 2 Diabetes, showing a robust relationship between anticipated and observed scores in both the training and testing sets. It revealed important patterns of connectivity in the anterior cingulate cortex and other regions responsible for cognitive control, which played a vital role in predicting cognitive deterioration. Our results indicate that machine learning approach, when using functional connectivity information, have the capability to forecast cognitive deterioration in T2D patients. This method may aid in early identification and intervention plans, potentially reducing the effects of cognitive deficits in this group. Further studies should confirm these results with larger and more varied samples to improve their applicability and relevance for clinical practice.

This study aims to develop a climate-resilient landscape design framework for coastal healthcare facilities by integrating Artificial Intelligence (AI)-generated design prompts with Computational Fluid Dynamics (CFD) simulations and on-site validation. Focusing on a coastal hospital in Penghu, Taiwan—a region vulnerable to strong winds, salt spray, and extreme weather—the research proposes a climate-adaptive, microclimate-responsive, and resilient design framework. Key findings demonstrate that the optimized design reduced average winter wind speed from 12 m/s to 4.5 m/s (a 62.5% reduction) and increased the three-year survival rate of salt-tolerant plant species (e.g., Pittosporum tobira, Casuarina) to 92%, significantly outperforming conventional planting strategies. The combination of water features and evapotranspiration planting reduced summer temperatures by 2.3 °C and increased humidity to 75%, with the PMV comfort index improving from +1.5 to +0.5. The program also resulted in a 15% increase in biodiversity, a 20% reduction in soil erosion, and a 40% improvement in users’ perceived aesthetic value of outdoor spaces. Furthermore, AI-based analyses to determine foundational depth led to a reduction in structural failure rates—from 40% to 5%—substantially elevating the safety and long-term durability of outdoor infrastructures. This study demonstrates that integrating AI with CFD is both feasible and highly effective for addressing complex coastal climate challenges in landscape architecture. The developed framework is parametric, evidence-based, and tailored to site-specific requirements, enabling the formulation of intelligent, climate-responsive landscape solutions for future healthcare environments in vulnerable coastal areas.

Adopting Martin Seligman’s PERMA model in education has been widely discussed in the literature. Still not enough, however, is known about it in the LOTE (Languages Other Than English) setting. This paper provides a detailed account of the deployment of PERMA in Polish language instruction in Taiwan. The lessons were conducted at the university and in the local Museum of World Religions (in New Taipei City). The author intended to see (a) the relationship between positive education and academic achievement, (b) how learners perceived religion in the curriculum, and (c) whether museum-based instruction enhanced learning. Given the soaring interest in PERMA, realistic accounts like this are essential. In this qualitative study, Seligman’s model provided a framework for exploring a potentially risky topic in an unobvious setting, a gallery filled with religious artifacts from all over the world. The data was obtained from a bilingual questionnaire and paired with the students’ actual performance, exam included. The instructor’s observations were described in detail. Unlike in other studies, the project did not produce excellent learning results, despite positive participant feedback and the teacher’s efforts to introduce and consolidate the material. One of the key findings was that a cheerful atmosphere of PERMA-enhanced learning did not necessarily translate into desired academic outcomes. Regarding religion in LOTE classrooms, there is no need to shy away from the topic if it is handled appropriately. The paper is meant to be a part of the global discussion on positive education while providing a realistic insight into the reality of foreign language instruction in this part of Asia.

The impedance analysis method, widely employed in human biology, is now gaining attention in biomedical research for a novel application: analyzing impedance signals during the egg incubation process to predict the gender of chickens within the eggs. In the contemporary landscape, Artificial Intelligence (AI) and Machine Learning play pivotal roles in livestock farming, offering cost‐effective and labor‐efficient solutions. In this study, 35 Taiwanese chicken eggs were used, comprising 20 male and 15 female eggs. During incubation, the impedance values of four electrodes (Z1, Z2, Z3, and Z4) were recorded. A constant voltage signal at 100 mV was applied, alternating between frequencies of 100 kHz and 8 MHz. The findings indicate that the impedance ratio between two pairs of electrodes, Z2 and Z3 compared to Z2 and Z4, on the eleventh day of incubation, reveals differences between male and female eggs within the frequency range of 1–2 MHz. The data set was trained using the Decision Tree approach in conjunction with k‐fold cross‐validation. The results demonstrate that the model achieved an accuracy of 81.48% in determining the gender of chicken eggs based on their impedance ratio. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

This study examines the effectiveness of augmented reality (AR) in digital learning by comparing three learning modalities: paper-based, computer-based, and AR-based presentations of artwork. A within-subject experimental design was employed with 10 participants, using a Latin Square Design to counterbalance the order of learning modalities and control for sequence effects. Each participant experienced all three modalities, followed by immediate and delayed recall tests to assess memory retention. The results will provide insights into how AR influences learning outcomes compared to traditional and digital methods. Findings from this study will contribute to the understanding of AR’s impact on memory performance and its potential applications in education.

Tetranychus ludeni and Frankliniella intonsa are voracious and economically important pests of crops. Orius strigicollis is a promising biocontrol agent for managing these pests, but effective integrated pest management (IPM) requires understanding factors that influence its predation efficiency. To investigate the impact of prey species, host plant, and predator sex, we measured trichome length and density on eggplant and bean leaves and assessed body dimensions and mobility of both predator and prey. Functional response analysis, Pearson’s correlation, and principal component analysis (PCA) were employed to determine the extent to which these factors affected handling time and attack rate under varying conditions. O. strigicollis consistently exhibited a Type II functional response when fed on both of thrip larvae and spider mites on eggplant and runner bean. Trichome length and density on leaves positively influenced predation by impeding mite movement, while deutonymphs of T. ludeni with minute body dimensions were more efficiently predated than 2nd instars of F. intonsa. Female predators demonstrated superior foraging performance compared to males, despite no observed differences in body size or mobility. These findings provide actionable insights for crop breeding and IPM strategies, and further field studies are recommended to validate these effects under varying environmental conditions.

Due to the advantages of low electrical resistance and size miniaturization, Cu-to-Cu direct bonding has become one of the important trends in microelectronic interconnect fabrication. Achieving robust Cu-to-Cu bonding at low temperature, ambient pressure, and with short processing time is crucial in practical applications. To fulfill these requirements, ultrasonic bonding techniques have been recently adopted. This study introduces a plasma-surface modification technique to ultrasonic Cu-to-Cu direct bonding, significantly enhancing bonding performance. Experimental results demonstrate that proper plasma bombardment can effectively increase joint strength. Remarkably, if the plasma treatment is applied exclusively to the upper bumps, the bonding performance can be further optimized, making the process more suitable for practical applications.