Department of Health Information Technology and Management, School of Allied Medical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
J Med Life. 2020 Oct-Dec;13(4):612-623. doi: 10.25122/jml-2020-0182.
The diagnosis of multiple sclerosis (MS) is difficult considering its complexity, variety in signs and symptoms, and its similarity to the signs and symptoms of other neurological diseases. The purpose of this study is to design and develop a clinical decision support system (CDSS) to help physicians diagnose MS with a relapsing-remitting phenotype. The CDSS software was developed in four stages: requirement analysis, system design, system development, and system evaluation. The Rational Rose and SQL Server were used to design the object-oriented conceptual model and develop the database. The C sharp programming language and the Visual Studio programming environment were used to develop the software. To evaluate the efficiency and applicability of the software, the data of 130 medical records of patients aged 20 to 40 between 2017 and 2019 were used along with the Nilsson standard questionnaire. SPSS Statistics was also used to analyze the data. For MS diagnosis, CDSS had a sensitivity, specificity and accuracy of 1, 0.97 and 0.99, respectively, and the area under the ROC curve was 0.98. The agreement rate of kappa coefficient (κ) between software diagnosis and physician's diagnosis was 0.98. The average score of software users was 98.33%, 96.65%, and 96.9% regarding the ease of learning, memorability, and satisfaction, respectively. Therefore, the applicability of the CDSS for MS diagnosis was confirmed by the neurologists. The evaluation findings show that CDSS can help physicians in the accurate and timely diagnosis of MS by using the rule-based method.
多发性硬化症（MS）的诊断较为复杂，其症状和体征多种多样，与其他神经疾病的症状和体征相似，因此诊断较为困难。本研究旨在设计和开发一种临床决策支持系统（CDSS），以帮助医生诊断具有复发缓解表型的 MS。CDSS 软件的开发分为四个阶段：需求分析、系统设计、系统开发和系统评估。使用 Rational Rose 和 SQL Server 设计面向对象的概念模型并开发数据库。使用 C sharp 编程语言和 Visual Studio 编程环境开发软件。为了评估软件的效率和适用性，使用了 2017 年至 2019 年间 130 名 20 至 40 岁患者的病历数据和 Nilsson 标准问卷。还使用 SPSS Statistics 分析数据。对于 MS 诊断，CDSS 的灵敏度、特异性和准确性分别为 1、0.97 和 0.99，ROC 曲线下面积为 0.98。软件诊断与医生诊断之间的 Kappa 系数（κ）一致性率为 0.98。软件用户在学习难易程度、可记性和满意度方面的平均得分分别为 98.33%、96.65%和 96.9%。因此，该 CDSS 得到了神经科医生的认可，其适用于 MS 的诊断。评估结果表明，CDSS 可以通过基于规则的方法帮助医生准确、及时地诊断 MS。
