Title
Application of patient-reported outcomes in clinical evaluation model of acupuncture for cervical spondylosis with SEM and ANNs
Abstract
Acupuncture treatment for Cervical Spondylosis (CS) has made great achievements at home and abroad. However its functions related patients' symptoms and treatment have not been assessed yet, which makes acupuncture difficult to be acknowledged by the authority of medical profession. Thankfully, patient reported outcome (PRO) scale is a qualified assessment tool originated from the West, which allows patients to describe subjective feelings and quality of life in self-reporting scales. The aim of this paper is to explore the applicability of PRO in clinical evaluation of acupuncture for CS and establish a clinical outcome evaluation model. Firstly, regular PRO scales (PROs) were introduced including SF-36 life quality questionnaire (SF-36) and visual analogue scale (VAS). To explore the applicability of PRO in clinical evaluation of acupuncture for CS, several statistical methods, such as reliability, validity, factor analysis and structural equation model (SEM), are fulfilled. As results, the value of Cronbach's statistic alpha of entire scale was 0.883, and 8 factors cumulative contribution rate achieved 70.59%, which illustrated the taxonomy of questions and the structure of the scale are acceptable. And then an evaluation index system was refined by revising scale with SEM. What's more, the study on a clinical outcome evaluation model of acupuncture for CS with ANN was carried out. BP neural network (BPNN) and Elman neural network were adopted to build models, and a better evaluation model was established by BPNN with an acceptable error of 0.88%, which outperformed Elman neural network. The experimental results indicate that PRO scales are suitable for clinical evaluation of acupuncture for CS and ANN is a promising method for establishing effective assessment model.
Year
DOI
Venue
2016
10.1109/ICMLC.2016.7872994
2016 International Conference on Machine Learning and Cybernetics (ICMLC)
Keywords
Field
DocType
Structural equation model,Artificial neural networks,Acupuncture,Cervical spondylosis(CS),Patient reported outcome (PRO)
Visual analogue scale,Quality of life,Structural equation modeling,Statistic,Computer science,Physical therapy,Cervical spondylosis,Artificial intelligence,Acupuncture,Cronbach's alpha,Machine learning,Patient-reported outcome
Conference
Volume
ISBN
Citations 
2
978-1-5090-0391-4
0
PageRank 
References 
Authors
0.34
1
7
Name
Order
Citations
PageRank
Li-Chun Guo100.34
Hong-Lai Zhang222.76
Hang Wei301.01
Xiao-Yan Wei400.34
Qin-Ye Lin500.34
Huo-Yuan Tan600.34
Qinqun Chen702.03