Affiliations: [a]
North China Coal Medical College Undergraduate, Tangshan, China
| [b]
National Institutes of Health, Bethesda, USA
Correspondence:
[*]
Corresponding author. Junwei Zhang, North China Coal Medical College Undergraduate, Tangshan 063000, China. Tel.: +86 0153731551539; E-mail: [email protected].
Note: [1] Yuxi Xie and Junwei Zhang contribute equally to this work.
Abstract: With the development of economic level and growth in the living standard, how to enhance the accuracy and efficiency of remote sepsis diagnosis has become the key issue and hot topic in the current medical research. As the traditional BP neural network is poor in the generalization ability and needs a large number of samples, we proposed the study on remote sepsis diagnosis by using the FOA-optimized BP neural network. Besides, based on the wireless sensor network and ZigBee network, we constructed the mathematical model of FOA_BP remote sepsis diagnosis based on wireless sensor network. The sepsis data set of School of Medicine and Public Health, University of Wisconsin was used as the object of study. The FOA-BP algorithm is significantly superior to the BP neural network algorithm in the diagnosis accuracy, verifying the validity and reliability of making a sepsis diagnosis with FOA_BP. Therefore, the method can be popularized to other fields to solve other similar problems.
Keywords: Wireless sensor network, ZigBee network, fruit fly algorithm, BP neural network, sepsis diagnosis