Decision making using electromyographic signals

使用肌电信号进行决策疾病肌肉的临床检查部分是通过对放置在收缩肌肉内的针电极获得的肌电数据的定性分析来执行的。虽然已经做出了一些努力来为这种分析提供量化框架，但这些定量技术在临床上的应用进展缓慢。这主要是因为临床工作人员不熟悉这些新技术，并且缺乏高质量的工具来帮助解释和探索这些数据。因此，我建议建立一种工具，旨在帮助临床专业人员探索特定患者的肌电数据，提供患者数据与各种类型肌肉疾病的相似性的上下文。为了实现这一目标，必须创建和验证规范和疾病数据的数据库，允许构建描述该数据的知识库。该知识库将通过在基于模式发现的规则系统中找到的应用统计分析技术来构建。使用如此构建的知识库，解释性决策支持工具将允许临床专业人员可视化地探索知识库内的关系，提供了一种明确的方法来建立接受检查的患者与描述先前检查显示肌肉疾病的不同状态和阶段的数据之间的关系。所描述的系统最初将侧重于肌病和神经病的疾病状态，以及辨别这些疾病状态和正常数据之间的差异的能力。在该数据库建立后，将有可能将该系统扩展到进一步的疾病类型，包括应激损伤和中风。

Decision Making using Electromyographic SignalsClinical examination of diseased muscle is performed, in part, through qualitative analysis of electromyographic data obtained from a needle electrode placed within a contracting muscle.While some efforts have been made to provide a quantitative framework for this analysis, clinical adoption of these quantitative techniques has been slow to occur.  This is largely due to the unfamiliarity of clinical staff with these new techniques, and a lack of quality tools to assist in the interpretation and exploration of this data.I therefore propose to establish a tool designed to assist a clinical professional in the exploration of the electromyographic data of a particular patient, providing a context of the similarity of the patient's data with the various types of muscular disease.  To achieve this goal, a database of normative and disease data must be created and validated, allowing the construction of a knowledge base describing this data.  This knowledge base will be constructed through the applied statistical analysis techniques found in the "pattern discovery" rule based system.Using the knowledge base so constructed, an interpretive decision support tool will allow the clinical professional to visually explore the relations within the knowledge base, providing a clear means of establishing the relationship between the patient under examination with the data describing prior examinations exhibiting various states and stages of muscular disease.The described system will initially focus on myopathic and neuropathic disease states, and the ability to discern the difference between these disease states and normal data.  Upon establishment of this database, it will be possible to extend the system to further disease types, including stress injury and stroke.