Neural nework-based mobile app framework to aid resource-poor setting community health

Residents who live in rural area normally have limited access to Internet, poor living condition and limited access to healthcare services. Besides, medical specialists are not available every day. As such, quality of healthcare services in rural area is extremely low. In Saudi Arabia, there are approximately 5.69 million people living in rural area (resource-poor settings). This work proposes a framework to develop mobile health application to aid the resource poor settings community health. The mobile health application makes use of the pre-trained neural network model to provide result of disease prediction for user. The application performance in term of accuracy, precision, recall and F1 score are measured using an existing diseases dataset and considering nine common diseases in the community. Measurement results show that the health mobile app developed thru the proposed framework is promising.