Urbanization trends in China reveal a shift in migration patterns, with an increasing number of recent migrants leaving primate cities while secondary cities emerge as attractive destinations. Given China's aging population and intensifying intercity competition for migrants, understanding the factors associated with recent migrants' intentions to leave cities at various levels becomes increasingly important. While spatial equilibrium theory implies that migrants' leaving intentions are shaped by the balance of housing, employment, and amenities, these factors vary hierarchically across city levels. This study examines how these factors differentially shape recent migrants' leaving intentions across primate, secondary, and small cities. Through binary logistic regression of national survey data, we find that recent migrants in secondary cities exhibit lower leaving intentions compared to those in primate and small cities. Further interaction analyses reveal distinct patterns: in primate cities, medium-income migrants are most likely to consider leaving; in secondary cities, rental housing status and hometown residential land ownership more strongly increase leaving intentions compared to primate cities, while medical resource accessibility more significantly reduces leaving intentions compared to small cities. Drawing on spatial equilibrium theory, our analysis suggests that secondary cities appear to achieve an optimal balance: their greater homeownership opportunities serve as a compensatory factor for their lower incomes compared to primate cities, while their superior medical amenity accessibility compensates for their higher housing costs compared to small cities. This paper contributes theoretically by bridging factors in spatial equilibrium theory with the urban hierarchy dynamics proposed by differential urbanization theory. It also offers practical insights for tailoring migration retention policies across city levels and adapting to the transformation of urbanization stages.

Evaluating the sustainable development level and obstacle factors of small towns is an important guarantee for implementing China's new-type urbanization and rural revitalization strategies, and is also a key path to promoting the United Nations Sustainable Development Goal 11 (SDG11). Traditional evaluation methods (such as Analytic Hierarchy Process, AHP, and Technique for Order Preference by Similarity to Ideal Solution, TOPSIS) mainly calculate the comprehensive score of each indicator through weighting. These methods have limitations in handling multidimensional data and system nonlinearity, and they cannot fully reveal the complex relationships and interactions within the sustainability systems of small towns. In contrast, the evaluation model combining Principal Component Analysis (PCA) and Catastrophe Progression Method (CPM) used in this study can better handle multidimensional data and system nonlinear relationships, reducing subjectivity in evaluation and improving the accuracy and reliability of the assessment results. The specific research process is as follows: First, based on the United Nations SDG11 framework, using multi-source big data, a theoretical framework and evaluation index system for the sustainable development of small towns suitable for the Chinese context were established. The impact of county-level factors on the sustainable development of small towns was also considered, and an entropy weight-grey correlation model was used to measure these impacts, resulting in a town-level dataset incorporating county-level influences. Secondly, the sustainability levels of 782 top small towns in China were evaluated using the comprehensive evaluation model based on PCA-CPM Model. Finally, an improved diagnostic model was used to identify obstacles influencing the sustainable development of small towns. The main findings include: 52.69% of the small towns have a sustainable development score exceeding 0.7255, indicating that the overall performance of small towns is at a medium to high development level. The development of small towns exhibits significant differences across regions and types, which are closely linked to county-level effects. Economic and social factors are the main obstacles to the sustainable development of small towns, and the impact of these obstacles intensifies from the eastern to the central, western, and northeastern regions. This study provides valuable insights for policymakers and scholars, promoting a deeper understanding of the sustainable development of small towns.

Exploring the influence of settlement patterns on the landscape fragmentation in woodlands and biological reserves is key to achieving ecologically sustainable development. In this research, we chose the Nanshan National Park in Hunan Province, China, as a case study, to explore the influence mechanisms. First, we identified the biological reserves through the landscape security patterns of biological conservation. Second, we constructed a coupling coordination model to analyze the coupling relationship between the settlement patterns and landscape fragmentation in the woodlands and biological reserves. The analysis showed that, overall, the effect of the settlement area on the landscape fragmentation in the biological reserves was more pronounced, while the effect of the settlement spread and shape on the landscape fragmentation in the woodlands was more obvious. From a type-specific perspective, we analyzed the coupling relationship between the settlement patterns and (1) the landscape fragmentation in different woodlands and (2) the landscape fragmentation in the biological reserves, namely concerning Leiothrix lutea and Emberiza aureola. We found that the effect of the settlement patterns on the landscape fragmentation of the Leiothrix lutea biological reserve was more significant than that of the landscape fragmentation of its main habitat, the evergreen broad-leaved forest. The effect of settlement patterns on the landscape fragmentation of the Emberiza aureola biological reserve was more significant than that of the landscape fragmentation of its other habitats. In addition, the results demonstrated that the habitat protection of the woodlands was not a substitute for the systematic protection of biosecurity patterns. This research could assist in developing more efficient conservation measures for ecologically protected sites with rural settlements.

Using short-wave infrared wavelength advantages, we demonstrate one-photon fluorescence confocal microscopy of adult mouse brains with penetration depths up to 1.7mm. This is achieved by labeling quantum dots with 1300 nm excitation and 1700 nm emission and detecting them with a single-photon superconducting nanowire detector.

With the advancement of urbanization, the stress on the green infrastructure around the urban agglomeration has intensified, which causes severe ecological problems. The uncertainty of urban growth makes it difficult to achieve effective protection only by setting protection red lines and other rigid measures. It is of practical significance to optimize the resilience of the stressed green infrastructure. To this end, we explore a scenario simulation analysis method for the resilience management of green infrastructure under stress. This research applies artificial neural network cellular automata to simulate the impacts of the Chang-Zhu-Tan urban agglomeration expansion on the green infrastructure in 2030 in three scenarios: no planning control, urban planning control, and ecological protection planning control. Based on the analysis, we identify four green infrastructure areas under stress and formulate resilience management measures, respectively. The results show that: (1) The distribution pattern of green infrastructure under stress is different in three scenarios. Even in the scenario of ecological protection planning and control, urban growth can easily break through the ecological protection boundary; (2) Residential, industrial, and traffic facility land are the main types of urban land causing green infrastructure stress, while forest, shrub, and wetland are the main types of the stressed green infrastructure; (3) Efficient protection of green infrastructure and the management of the urban growth boundary should be promoted by resilient management measures such as urban planning adjustment, regulatory detailed planning, development strength control and setting up the ecological protection facilities for the stressed green infrastructure areas of the planning scenarios and the no-planning control scenarios, for the areas to be occupied by urban land, and for the important ecological corridors. The results of this study provide an empirical foundation for formulating policies and the methods of this study can be applied to urban ecological planning and green infrastructure management practice in other areas as well.

Outdoor physical activities can promote public health and they are largely influenced by the built environment in different urban settings. Understanding the association between outdoor physical activities and the built environment is important for promoting a high quality of life. Existing studies typically focus on one type of outdoor activity using interview-based small samples and are often lack of systematic understanding of the activities' intensity and frequency. In this study, we intend to gain deeper insight into how the built environment influences physical activities using the data extracted from individual's wearables and other open data sources for integrated analysis. Multi-linear regression with logarithm transformation is applied to perform the analysis using the data from Changsha, China. We found that built environment impacts on outdoor physical activities in Changsha are not always consistent with similar studies' results in other cities. The most effective measures to promote outdoor physical activities are the provision of good arterial and secondary road networks, community parks, among others in Changsha. The results shed light on future urban planning practices in terms of promoting public health

Analysis of longitudinal changes in imaging studies often involves both segmentation of structures of interest and registration of multiple timeframes. The accuracy of such analysis could benefit from a tailored framework that jointly optimizes both tasks to fully exploit the information available in the longitudinal data. Most learning- based registration algorithms, including joint optimization approaches, currently suffer from bias due to selection of a fixed reference frame and only support pairwise transformations. We here propose an analytical framework based on an unbiased learning strategy for group-wise registration that simultaneously registers images to the mean space of a group to obtain consistent segmentations. We evaluate the proposed method on longitudinal analysis of a white matter tract in a brain MRI dataset with 2-3 time-points for 3249 individuals, i.e., 8045 images in total. The reproducibility of the method is evaluated on test-retest data from 97 individuals. The results confirm that the implicit reference image is an average of the input image. In addition, the proposed framework leads to consistent segmentations and significantly lower processing bias than that of a pair-wise fixed-reference approach. This processing bias is even smaller than those obtained when translating segmentations by only one voxel, which can be attributed to subtle numerical instabilities and interpolation. Therefore, we postulate that the proposed mean-space learning strategy could be widely applied to learning-based registration tasks. In addition, this group-wise framework introduces a novel way for learning-based longitudinal studies by direct construction of an unbiased within-subject template and allowing reliable and efficient analysis of spatio-temporal imaging biomarkers.

Optical microscopy is a valuable tool for in vivo monitoring of biological structures and functions because of its noninvasiveness. However, imaging deep into biological tissues is challenging due to the scattering and absorption of light. Previous research has shown that the two optimal wavelength windows for high-resolution deep mouse brain imaging are around 1300 and 1700 nm. However, one-photon fluorescence imaging in the wavelength region has been highly challenging due to the poor detection efficiency of currently available detectors. To fully utilize this wavelength advantage, we demonstrated here one-photon confocal fluorescence imaging of deep mouse brains with an excitation wavelength of 1310 nm and an emission wavelength within the 1700 nm window. Fluorescence emission at 1700 nm was detected by a custom-built superconducting nanowire single-photon detector (SNSPD) optimized for detection between 1600 nm and 2000 nm with low detection noise and high detection efficiency. With the PEGylated quantum dots and SNSPD both positioned at the optimal imaging window for deep tissue penetration, we demonstrated in vivo one-photon confocal fluorescence imaging at approximately 1.7 mm below the surface of the mouse brain, through the entire cortical column and into the hippocampus region with a low-cost continuous-wave laser source and low excitation power. We further discussed the significance of the staining inhomogeneity in determining the depth limit of one-photon confocal fluorescence imaging. Our work may motivate the further development of long wavelength fluorescent probes, and inspire innovations in high-efficiency, high-gain, and low-noise long wavelength detectors for biological imaging.

This work presents a single-step deep-learning framework for longitudinal image analysis, coined Segis-Net. To optimally exploit information available in longitudinal data, this method concurrently learns a multi-class segmentation and nonlinear registration. Segmentation and registration are modeled using a convolutional neural network and optimized simultaneously for their mutual benefit. An objective function that optimizes spatial correspondence for the segmented structures across time-points is proposed. We applied Segis-Net to the analysis of white matter tracts from N=8045 longitudinal brain MRI datasets of 3249 elderly individuals. Segis-Net approach showed a significant increase in registration accuracy, spatio-temporal segmentation consistency, and reproducibility compared with two multistage pipelines. This also led to a significant reduction in the sample-size that would be required to achieve the same statistical power in analyzing tract-specific measures. Thus, we expect that Segis-Net can serve as a new reliable tool to support longitudinal imaging studies to investigate macro- and microstructural brain changes over time.

Confounding bias is a crucial problem when applying machine learning to practice, especially in clinical practice. We consider the problem of learning representations independent to multiple biases. In literature, this is mostly solved by purging the bias information from learned representations. We however expect this strategy to harm the diversity of information in the representation, and thus limiting its prospective usage (e.g., interpretation). Therefore, we propose to mitigate the bias while keeping almost all information in the latent representations, which enables us to observe and interpret them as well. To achieve this, we project latent features onto a learned vector direction, and enforce the independence between biases and projected features rather than all learned features. To interpret the mapping between projected features and input data, we propose projection-wise disentangling: a sampling and reconstruction along the learned vector direction. The proposed method was evaluated on the analysis of 3D facial shape and patient characteristics (N = 5011). Experiments showed that this conceptually simple method achieved state-of-the-art fair prediction performance and interpretability, showing its great potential for clinical applications.