mHealth Technology Translation in a Limited Resources Community-Process, Challenges, and Lessons Learned From a Limited Resources Community of Chiang Mai Province, Thailand

Regional maldistribution of human resources of rehabilitation institutions in China Mainland based on spatial analysis

Objective

With the demand for rehabilitation has been increasing faster for the aging of China's population, the equity of rehabilitation resource has aroused great concern. This study aimed to analyze the spatial distribution and evolution of the human resources of rehabilitation institutions and propose targeted countermeasures and suggestions to promote optimal allocation.

Methods

A total of 31 provinces in China Mainland were identified and geocoded. The spatial weight matrix was introduced to analyze the spatial correlation. Spatial autocorrelation analysis and tests were used to analyze the spatial distribution and evolution characteristics of rehabilitation institutions' human resources in China from 2016 to 2019.

Results

The average density of rehabilitation staff from 2016 to 2019 has been rising yearly (From 1.60 to 1.88). From 2018 to 2019, the proportion of rehabilitation professionals was higher than 75% in only 5 provinces, and no provinces met 75% from 2016 to 2017. Global Moran's I index from 2016 to 2019 showed no apparent aggregation phenomenon in the allocation of management personnel resources (P > 0.05). Three provinces in western China belonged to the Low-Low area and a province in northeastern China fitted to the Low-High area, with statistically significant differences. In addition, the changes in the spatial distribution and evolution trend of the human resources of rehabilitation institutions in different periods were affected by health policies.

Conclusions

Although the overall spatial distribution gap of human resource allocation of rehabilitation institutions is shrinking, there are still internal structural defects and a maldistribution at the provincial level. It is necessary to improve the overall number of staff in rehabilitation institutions and to ameliorate the proportion of different types of staffing.

Influenza, dengue and common cold detection using LSTM with fully connected neural network and keywords selection

Symptom-based machine learning models for disease detection are a way to reduce the workload of doctors when they have too many patients. Currently, there are many research studies on machine learning or deep learning for disease detection or clinical departments classification, using text of patient’s symptoms and vital signs. In this study, we used the Long Short-term Memory (LSTM) with a fully connected neural network model for classification, where the LSTM model was used to receive the patient’s symptoms text as input data. The fully connected neural network was used to receive other input data from the patients, including body temperature, age, gender, and the month the patients received care in. In this research, a data preprocessing algorithm was improved by using keyword selection to reduce the complexity of input data for overfitting problem prevention. The results showed that the LSTM with fully connected neural network model performed better than the LSTM model. The keyword selection method also increases model performance.