Voice Based, Workflow Enhancing, Primary Care Medical Data Input System

描述(由申请人提供)：在医院和医生办公室采用电子健康记录(EHR)作为解决各种医疗保健问题的单一解决方案已得到广泛推广。然而，美国84%的中小型企业(SMB)医生诊所尚未采用EHR系统。介入动力学公司(IDC)已经进行了200多次初级保健医生访谈，发现阻碍采用的主要因素是工作流程延迟和费用。降低工作流速度的最大因素是数据输入过程。IDC提出的项目有一个明确的目标：利用创新的语音输入技术和开放源代码系统开发一种低成本的自动化解决方案，使初级保健医生能够在患者检查过程中完整地完成初级保健笔记。叙述性语音输入将以上下文敏感、领域受限的方式进行分析，以产生结构化的临床数据，这些数据可以很容易地集成到符合标准的电子病历中。通过使用直接转换为相关EHR条目的语音输入，医生可以提高他们笔记的准确性，消除第三方转录错误，并避免工作流程延迟。项目方法将包括： 进一步测试和最终开发IDC正在申请专利的语音系统DocTalk，该系统可以对结构化医疗信息进行准确的自然语言处理；开发一个概念验证数据系统，使用领域增强的开放源代码将医生的语音输入从语音到文本再到结构化文本再到电子病历数据；评估概念验证系统相对于传统电子病历输入方法的有效性，目标如下：实现绘图时间减少50%或更多，输出准确率达到90%或更高，以及在包括可学习性、工作流程匹配、可用性和总体满意度在内的主观指标上获得大于4分的分数。拟议计划的成功完成将为IDC提供一个可行的技术平台，该平台可以立即对初级保健医生生成结构化文档以用于其当前的EHR平台有用。此外，在该计划内开发和改进的技术可以以多种方式扩展。 与公共健康相关：IDC技术旨在绕过中小企业市场采用的正常障碍，以最低的价格点快速提高工作流程和患者护理质量。IDC将为目前使用笔和纸的医生提供更自然、更快速的临床数据输入方式，消除花费在键盘输入或复杂的电子病历屏幕导航上的时间。该系统将生成结构化的临床数据，以实现健康信息的交换、患者记录的可携带性、帐单、数据分析(包括当地执业和公共卫生)、营销和其他好处，从而降低总体医疗成本。

DESCRIPTION (provided by applicant): The adoption of electronic health records (EHR) in hospitals and physician offices has been widely promoted as a single solution to a wide variety of health care issues. Yet 84% of small and medium business (SMB) physician practices in the US have not adopted EHR systems. Interventional Dynamics Corporation (IDC) has conducted more than 200 primary care physician interviews, finding that the major disincentives to adoption are workflow delay and expense. The single greatest factor in the reduction of workflow speed is the data input process. IDC's proposed project has this specific aim: Utilize an innovative voice entry technique and open source code systems to develop a low-cost, automated solution to allow primary care physicians to complete a primary care note entirely during the patient examination process. The narrative speech input will be analyzed in a context-sensitive, domain-restricted manner to produce structured clinical data that can be readily integrated into standards-compliant electronic medical records. By using speech inputs that are converted directly to relevant EHR entries, physicians can increase the accuracy of their notes, eliminate third party transcription errors and avoid workflow delays. The project approach will include: Further testing and final development of DocTalk, the IDC patent pending speech system that allows accurate natural language processing of structured medical information; Development of a proof-of-concept data system that converts physician voice input from voice to text to structured text to EHR data using domain enhanced open source code; The evaluation of the effectiveness of the proof-of-concept system against traditional EHR input methods with the following goals: Achieve 50% or more reduction in charting time, achieve 90% or more accuracy in output, and score greater than 4 of 5 on subjective metrics including learnability, workflow fit, usability, and overall satisfaction. Successful completion of the proposed program will provide IDC with a viable technology platform that can immediately be useful to primary care physicians in generating structured documents for use with their current EHR platforms. Furthermore, the technology developed and refined within this program can be expanded in multiple ways. PUBLIC HEALTH RELEVANCE: The IDC technology is designed to circumvent the normal barriers to adoption in the SMB market and allow for quick increases in workflow and quality of patient care at a minimal price point. IDC will provide physicians who currently use pen and paper a more natural and faster way to input clinical data, eliminating time spent on hunt-and-peck keyboard entry or complicated EHR screen navigation. The system will generate structured clinical data that enables the exchange of health information, the portability of patient records, billing, data analytics (both local practice and public health), marketing, and other benefits, resulting in the reduction of overall healthcare costs.