A clinical decision support system for the diagnosis of probable migraine and probable tension-type headache based on case-based reasoning

The Journal of Headache and Pain

Research Article

Xudong Lu¹  Ziming Yin¹  Huilong Duan¹  Xiaoyan Chen²  Shengyuan Yu²  Zhao Dong² 

[1] College of Biomedical Engineering and Instrument Science, Zhejiang University, Zheda Road 38, 310008, Hangzhou, Zhejiang, China;International Headache Center, Department of Neurology, Chinese PLA General Hospital, 100853, Beijing, China;

关键词: Headache;    Clinical decision support;    Case-based reasoning;   

DOI  :  10.1186/s10194-015-0512-x

 received in 2015-01-12, accepted in 2015-03-15,  发布年份 2015

来源: Springer

【 摘 要 】

BackgroundThe overlap between probable migraine (PM) and probable tension-type headache (PTTH) often confuses physicians in clinical practice. Although clinical decision support systems (CDSSs) have been proven to be helpful in the diagnosis of primary headaches, the existing guideline-based headache disorder CDSSs do not perform adequately due to this overlapping issue. Thus, in this study, a CDSS based on case-based reasoning (CBR) was developed in order to solve this problem.MethodsFirst, a case library consisting of 676 clinical cases, 56.95% of which had been diagnosed with PM and 43.05% of which had been diagnosed with PTTH, was constructed, screened by a three-member panel, and weighted by engineers. Next, the resulting case library was used to diagnose current cases based on their similarities to the previous cases. The test dataset was composed of an additional 222 historical cases, 76.1% of which had been diagnosed with PM and 23.9% of which had been diagnosed with PTTH. The cases that comprised the case library as well as the test dataset were actual clinical cases obtained from the International Headache Center in Chinese PLA General Hospital.ResultsThe results indicated that the PM and PTTH recall rates were equal to 97.02% and 77.78%, which were 34.31% and 16.91% higher than that of the guideline-based CDSS, respectively. Furthermore, the PM and PTTH precision rates were equal to 93.14% and 89.36%, which were7.09% and 15.68% higher than that of the guideline-based CDSS, respectively. Comparing CBR CDSS and guideline-based CDSS, the p-value of PM diagnoses was equal to 0.019, while that of PTTH diagnoses was equal to 0.002, which indicated that there was a significant difference between the two approaches.ConclusionsThe experimental results indicated that the CBR CDSS developed in this study diagnosed PM and PTTH with a high degree of accuracy and performed better than the guideline-based CDSS. This system could be used as a diagnostic tool to assist general practitioners in distinguishing PM from PTTH.

【 授权许可】

Unknown   
© Yin et al.; licensee Springer. 2015. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.