No abstract

The aging world population requires a sustainable and high-quality healthcare system. To examine the efficiency of medical cooperation, medical provider and physician networks were constructed using patient claims data. Previous studies have shown that these networks contain information on medical cooperation. However, the usage patterns of multiple medical providers in a series of medical services have not been considered. In addition, these studies used only general network features to represent medical cooperation, but their expressive ability was low. To overcome these limitations, we analyzed the medical provider network to examine its overall contribution to the quality of healthcare provided by cooperation between medical providers in a series of medical services. This study focused on: i) the method of feature extraction from the network, ii) incorporation of the usage pattern of medical providers, and iii) expressive ability of the statistical model. Femoral neck fractures were selected as the target disease. To build the medical provider networks, we analyzed the patient claims data from a single prefecture in Japan between January 1, 2014 and December 31, 2019. We considered four types of models. Models 1 and 2 use node strength and linear regression, with Model 2 also incorporating patient age as an input. Models 3 and 4 use feature representation by node2vec with linear regression and regression tree ensemble, a machine learning method. The results showed that medical providers with higher levels of cooperation reduce the duration of hospital stay. The overall contribution of the medical cooperation to the duration of hospital stay extracted from the medical provider network using node2vec is approximately 20%, which is approximately 20 times higher than the model using strength.

As social behavior plays an essential role in people's lives, the features of face-to-face interaction networks must be examined to understand people's social behavior. In this study, we focused on the stable community structure of a faceto-face interaction network because it explains the persistent communities caused by the stationary communication patterns of citizens and visitors in a city. We regarded citizens and visitors as two kinds of particles and the community as a phase and theorized the stability of the community structure using the equilibrium conditions among communities. We formulated the chemical potentials of the communities and examined whether they were in equilibrium under the assumption of a canonical ensemble. We estimated the chemical potentials of persistent communities and found that these values matched within approximately 10% error for each day. This result indicates that the cause of persistent communities is the stability of community structure.

Social connections among people are mediated by multiple means of communication, including face-to-face interaction, cell phones, and social networking services. Therefore, to understand people's social behavior, it is necessary to integrate social networks using multiple means of communication and to understand them comprehensively. Large-scale face-to-face interaction networks have not been constructed in previous studies. Thus, we develop a method to construct a large-scale face-to-face interaction network using mobility data. Then, we calculate the chemical potential of each community using the spatial coordinates of the constructed network and investigate the stability of the community structure from a statistical mechanical point of view. As a result, we constructed a face-to-face interaction network with about 5,000 people using 39 days of data in Kyoto City. We confirmed that the network structure for the entire period was robust and had a similar geographic space allocation of communities. The chemical potentials of the communities in the constructed network were calculated, and the error of the chemical potentials among the communities was about 10%. We investigated the contribution of the chemical potential from the ideal gas and the chemical potential from the particle interaction. And we consider that the stability of the community structure of the face-to-face interaction network is determined by the variation of the particle density in each community and the localization of the particle distribution in the community.

The aging world population requires a sustainable and high-quality healthcare system. To examine the efficiency of medical cooperation, medical provider and physician networks were constructed using patient claims data. Previous studies have shown that these networks contain information on medical cooperation. However, the usage patterns of multiple medical providers in a series of medical services have not been considered. In addition, these studies used only general network features to represent medical cooperation, but their expressive ability was low. To overcome these limitations, we analyzed the medical provider network to examine its overall contribution to the quality of healthcare provided by cooperation between medical providers in a series of medical services. This study focused on: i) the method of feature extraction from the network, ii) incorporation of the usage pattern of medical providers, and iii) expressive ability of the statistical model. Femoral neck fractures were selected as the target disease. To build the medical provider networks, we analyzed the patient claims data from a single prefecture in Japan between January 1, 2014 and December 31, 2019. We considered four types of models: a model using node strength and linear regression to a model using feature representation by node2vec and regression tree ensemble, which is a machine learning method. The results showed that a stronger medical provider reduces the duration of hospital stay. The overall contribution of the medical cooperation to the duration of hospital stay extracted from the medical provider network using node2vec is approximately 20%, which is approximately 20 times higher than the model using strength.