Research by our members

Rapid urbanization has significantly intensified the impact of natural disasters, particularly floods, posing a severe threat to urban safety. To enhance urban flood risk management, a quantitative assessment of urban flood vulnerability is essential. This study integrates multi-source data to develop the urban Development Index (DI) and the Normalized Difference Frequency Index (NDFI). By applying area-weighted aggregation, spatial overlay analysis and K-Means clustering, this research explores the spatiotemporal dynamics of urban expansion and its relationship with flood vulnerability in the Chengdu-Chongqing urban agglomeration. The results indicate that between 1995 and 2020, the region experienced significant urban expansion, with urban land increasing by 4,232 km2. Moreover, the newly developed urban areas were found to be more vulnerable to flooding, with 42.26 % affected by floods in 2020. However, the flood vulnerability did not show a simple linear relationship with development level. In more developed cities, such as Chongqing, showed decreasing vulnerability (from level C to A) due to high-elevation terrain and sustained investment in flood mitigation. In contrast, Chengdu experienced increasing vulnerability (from level A to C) due to rapid expansion over flat agricultural land with dense construction. Furthermore, Ziyang showed high flood vulnerability despite slow development, with over 80 % of its newly developed urban areas were affected by floods. The proposed flood vulnerability assessment framework provides a scientific basis for flood risk management in Chengdu-Chongqing urban agglomeration and other similar inland city clusters. Additionally, it serves as a valuable reference for future urban expansion planning and disaster prevention strategies.

Background Intergenerational transmission of physical abuse is well-known; however, the potential for contextual effect modification remains unknown. Objective We investigated whether the norms of tolerance for child physical abuse modify the intergenerational transmission of physical abuse. Participants and setting Data from the Child Living Standard Survey (grades five and eight) conducted between 2016 and 2018 in three prefectures in Japan was analyzed. Methods The norms of tolerance for child physical abuse were calculated based on the proportion of caregivers who physically abuse their offspring in each of the 33 districts (n = 43,534). The caregivers' childhood physical abuse victimization and perpetration of child physical abuse were assessed via questionnaire. A multilevel analysis was performed to evaluate the cross-level interaction of the norms of tolerance for child physical abuse at the neighborhood level on the association between the caregiver's childhood physical abuse victimization and perpetration of child physical abuse, stratified by fathers (n = 4334) and mothers (n = 38,290). Results By district, the average percentage of caregivers who physically abuse their offspring was 14.4 %. Intergenerational transmission was confirmed for both fathers and mothers. Fathers who are living with highly tolerant norms for child physical abuse showed higher odds of intergenerational transmission of physical abuse, while mothers showed no effect modification. Conclusion The norms of tolerance for child physical abuse showed a more prominent intergenerational transmission of physical abuse for fathers but not for mothers. To modify the cycle of physical abuse, interventions to reduce norms of tolerance for child physical abuse may be useful.

Deep learning has become an essential tool for identifying transcription factor (TF) binding sites, yet conventional approaches often struggle with limited training data for specific TFs. Here, we introduce QTFPred (Quantum-based TF Predictor), a quantum-classical hybrid framework that integrates quantum convolutional layers within neural networks to predict TF binding at base resolution. By leveraging the exponential feature space offered by quantum circuits and training from scratch via GPU simulation, QTFPred achieves robust performance even in data-sparse scenarios. In benchmarks on 49 Encyclopedia of DNA elements ChIP-seq datasets, QTFPred delivered state-of-the-art accuracy in 92% of binary prediction and 96% of signal prediction tasks, outperforming conventional models in precision and stability. Moreover, the method reveals underlying TF motif representations, offering insights into cooperative binding mechanisms. These results highlight the potential of quantum machine learning to overcome the limitations of traditional deep learning in genomics modeling.

Background While Japan provides universal healthcare, immigrants may experience hampered access to healthcare. A comprehensive review of immigrant healthcare access is also lacking. This systematic review aims to examine barriers and facilitators of healthcare access among immigrants in Japan. Methods We searched for literature published in English and Japanese until January 9, 2024. Studies were included if they assessed factors influencing any stage of immigrants’ healthcare access, such as perceiving needs, seeking, reaching, utilizing healthcare and the consequences of healthcare, as defined by the Levesque framework. We performed a thematic analysis to further identify categories (PROSPERO: CRD42023418554). Findings After screening 2791 articles, we identified 67 studies (40 quantitative, 23 qualitative, 4 mixed methods) meeting eligibility criteria. Limited healthcare information led immigrants to seek alternative information sources and affected immigrants' perceived healthcare needs. Longer duration of stay improved access to healthcare information. Cultural and healthcare system differences affected healthcare seeking. Reaching and utilizing healthcare were hindered by heavy workloads, undocumented status, financial hardship, and limited insurance but were facilitated by support from family and friends. The healthcare system was often insufficient to support immigrants’ language and cultural needs leading to dissatisfaction and poor compliance. Interpretation Findings highlight the critical importance of a multidimensional approach to support immigrants in Japan, ranging from improving healthcare information access to creating immigrant-friendly health systems. More research is needed on the healthcare access among vulnerable immigrants, such as undocumented and low-skilled labor immigrants and children. Funding No specific funding source supported this study.

Land use and land cover (LULC) transformations significantly impact ecosystem hydrological characteristics. This study analysed the Sakumo catchment from 1990 to 2020 to assess changes in runoff volume and nonpoint source (NPS) pollutant loads. Using remote sensing, GIS products, and the QNSPECT (QGIS Plugin for Nonpoint Source Pollution and Erosion Comparison Tool) hydrologic model, we quantified and spatially analysed rainfall-runoff and pollutant loads. Our results revealed increasing trends in runoff (110.3 mm/year) and pollutant loads (total nitrogen: 0.22 kg/year; total phosphorus: 0.34 kg/year), attributed to population growth (1, 475 ppl/km²) and urbanization-induced LULC changes (Development: 44.7%). GIS outputs identified the sources of these increments. This research addresses critical gaps in understanding Ghanaian wetlands by estimating runoff magnitudes, identifying spatial pollution sources, and assessing catchment-scale impacts on wetland functions. The study's findings align with historical water quality trends in the Sakumo Ramsar Site, and our runoff estimates demonstrate satisfactory agreement with global runoff model products. This study provides a practical approach for stakeholders in data-deficient regions to investigate hydrological and environmental changes, guiding informed decision-making.

Coastal wetlands are complex ecosystems that support biodiversity. They provide many benefits, including flood mitigation and sustenance for communities. The unique characteristics of wetlands make them vulnerable to natural and human-induced disturbances. Numerous factors, including industrialisation, urbanisation, and climate change, add to this phenomenon. The activities that threaten coastal wetlands in the world are relevant to coastal wetlands in Ghana. The Songor and Sakumo wetland catchments are international ecosystems endangered by land modifications and sea level rise. There are gaps in the body of knowledge that need investigation as regards underlying processes and transformation. This study assessed land use and land cover (LULC) changes between 1990 and 2020. The study used geospatial techniques and intensity analysis. LULC change results were from Landsat images (1990, 2000, 2011, and 2020). These changes were attributed to an increase in human activities. Changes in the Sakumo wetland catchment fell more into human-induced LULC categories, and vice versa for the Songor wetland catchment. The study recommends comprehensive methods of LULC change analysis. This would enhance biodiversity and allow the sustainable usage of wetland resources.