Park green space is an important part of the green infrastructure in a city, and urban park space accessibility is an important indicator for measuring the city’s comprehensive strength, living environment, and resident happiness, which greatly affects the sustainable development and construction of the city. The spatial syntax method can discuss the accessibility of urban parks from the perspective of topology, which inherits the objective stability of the topological relationship. There are few research studies on the accessibility of urban parks. Therefore, with the support of space syntax theory and the spatial statistics method, this paper combined the data of park green space and road network to complete the accessibility evaluation of urban parks in the Changsha City urban area. This paper evaluates the accessibility of urban parks in the study area from four aspects, namely global accessibility, perceptual accessibility, local accessibility, and psychological accessibility, by using five quantitative indexes of space syntax, namely connectivity, depth, integration, selection, and synergy. The conclusions are as follows: The spatial layout area of the park is roughly consistent with the areas that have good global and local accessibility of the road network, and the global accessibility and local accessibility of the park in the study area are relatively high. The global spatial structure can be better perceived by the local space at a radius scale greater than or equal to 3000 m, and the perceptual accessibility of the park is high. Most of the urban parks in the Changsha urban area are located in sections with high space efficiency. The park space penetration is good, and nearly 70% of the parks have high psychological accessibility. Based on the overall spatial layout of urban parks, the space syntax method quantitatively and comprehensively evaluates the accessibility of urban parks from the perspective of topology. The conclusion is reliable and has important application value in the evaluation of park accessibility. It can be used as an important supplement to improve the evaluation model of accessibility.

For Case-II water bodies with relatively complex water qualities, it is challenging to establish a chlorophyll-a concentration (Chl-a concentration) inversion model with strong applicability and high accuracy. Convolutional Neural Network (CNN) shows excellent performance in image target recognition and natural language processing. However, there little research exists on the inversion of Chl-a concentration in water using convolutional neural networks. Taking China’s Dongting Lake as an example, 90 water samples and their spectra were collected in this study. Using eight combinations as independent variables and Chl-a concentration as the dependent variable, a CNN model was constructed to invert Chl-a concentration. The results showed that: (1) The CNN model of the original spectrum has a worse inversion effect than the CNN model of the preprocessed spectrum. The determination coefficient (R_P²) of the predicted sample is increased from 0.79 to 0.88, and the root mean square error (RMSE_P) of the predicted sample is reduced from 0.61 to 0.49, indicating that preprocessing can significantly improve the inversion effect of the model.; (2) among the combined models, the CNN model with Baseline1_SC (strong correlation factor of 500–750 nm baseline) has the best effect, with R_P² reaching 0.90 and RMSEP only 0.45. The average inversion effect of the eight CNN models is better. The average R_P² reaches 0.86 and the RMSE_P is only 0.52, indicating the feasibility of applying CNN to Chl-a concentration inversion modeling; (3) the performance of the CNN model (Baseline1_SC (R_P² = 0.90, RMSE_P = 0.45)) was far better than the traditional model of the same combination, i.e., the linear regression model (R_P² = 0.61, RMSE_P = 0.72) and partial least squares regression model (Baseline1_SC (R_P² = 0.58. RMSE_P = 0.95)), indicating the superiority of the convolutional neural network inversion modeling of water body Chl-a concentration.

Background: Practical implementation of radiomics research faces significant data accessibility challenges due to privacy and ethical restrictions on multicenter data aggregation. Federated Learning (FL) provides a secure distributed framework that preserves data privacy through cryptographic techniques. Its adoption in radiomics is an emerging trend, enabling collaborative training without sharing sensitive imaging data. However, the inherently Non-IID data distribution across clients in FL often leads to class imbalance, which can substantially degrade global model performance. Purpose: To develop a privacy-preserving, multicenter collaborative CT-radiomics model for evaluating neoadjuvant chemotherapy efficacy in non‑small cell lung cancer (NSCLC). Methods: To mitigate FL performance degradation caused by data imbalance, we propose a parameter‑sharing federated aggregation algorithm (FedPS), where model parameters are sequentially shared via the server. Results: On an imbalanced NSCLC NAC efficacy dataset, centralized learning achieved an AUC of 0.92. FedPS attained competitive performance (AUC = 0.88), approaching the centralized benchmark while preserving privacy. Common FL algorithms performed lower: FedAvg (AUC = 0.84), FedSGD (0.85), and FedProx (0.85). On extremely imbalanced data, FedPS maintained good performance (AUC = 0.86), compared to FedAvg (0.80), FedSGD (0.83), and FedProx (0.85). Conclusions: The proposed FedPS algorithm demonstrates promising classification and generalization performance in imbalanced federated learning scenarios.