Eyes on FHIR
什么是 Eyes on FHIR ?
从事医疗信息化的朋友可能听说过FHIR(Fast Healthcare Interoperability Resouce), 但是在具体实现和应用上可能没有相关经验。下面这篇文章里,我们一起来看看美国的一个Patient Care project在眼科是怎么落地FHIR的。
背景
2020年美国《21世纪医疗法》最终法规要求健康技术供应商必须支持2个公开的API能够使患者访问自己的健康数据。这两个API需要遵循HL7 FHIR标准。随着社区力量的驱动,FHIR的流行度和成熟度在美国逐渐成型,使得健康信息数据互操作性得到了很大的提升。
Eyes on FHIR的目的
虽然在美国FHIR的发展有很大进展,但是真实世界项目的落地尤其是临床专科的应用仍十分有限,主要原因是没有统一的实施指引(Implementation Guidance)。所以这个项目的目标就是开发必要的国际标准使FHIR能够真正地给某个专科带来价值。此项目的实施方式主要通过把眼科尽量全面的临床词典映射到FHIR格式,最后形成实施标准来帮助实现系统间的互联互通。
具体流程如下:
列出一系列眼科真实世界的用例,尤其是有互操作性痛点的案例。用例的形式需要考虑如下情形:
- 涉及的元素
- Actors 操作人
- Pre-condition 先决条件
- Workflow 工作流
- Post-condition 后置条件
- Exceptions 例外
- Data elements 数据元素
- Necessary commentary 必要的备注
- 专用词汇的定义
- FHIR 资源
- 临床专业术语, 如SNOMED-CT, LOINC, DICOM
- 局限和差距的发现
- 临床上,在某些地方的设备代码并不全面
- 如何做FHIR profile
- 涉及的元素
收集和整理如上信息,包括编码系统等。
把上述信息转化为FHIR Profiles, 配置是否是必须项。
开发实施指引IG, 甄别如何使用FHIR资源解决某个特定的互操作性问题。
发布实施指引
参加三年一次的技术连接会,演示用例和测试互操作性。
参加三年一次的HL7-FHIR投票,获准正式发布实施指引。
转化IG, 让更多真实世界的案例可以使用此指引。
项目预期结果
- 建立治理、质控和伦理的原则;促成眼科相关各界人事代表组成委员会共同发展未来路线图。
- 临床和研究影响力。作为第一个临床专科实施项目,藉此机会展示FHIR的临床好处;记录可以重复利用的技术、协议、流程等材料。
此项目使用的场景
- 日常临床关护沟通相关;病人旅程相关,比如:纵向病患旅程;异步通信;远程医疗设备等
- 医师临床工具, 如SMART app/CDS hooks/integrating remote monitoring
- 患者注册数据收集, 如批量数据导出
- 临床工作流优化,如自动支付、审计、优先授权、行政管理。
- 辅助生命科学和研发,如辅助临床招募,数据统一。
- AI图像诊断工作流,如下图:
具体细节功能
- Providing guidance for the FHIR representation of a comprehensive set of codified clinical findings to facilitate interoperable exchange of these data elements between systems.
- Providing guidance for the FHIR representation of a comprehensive set of codified diagnoses to facilitate interoperable exchange of these data elements between systems.
- Providing guidance for the FHIR representation of a comprehensive set of codified retinal therapeutics to facilitate interoperable exchange of these data elements between systems.
- Exchanging information between and EHR (/PMS) and a diagnostic device where:
EHR Supports DICOM Modality Worklist, Storage and Display (With no PACS)
Based on IHE’s Unified Eyecare Workflow ‘real world model’ (RWM) II - Exchanging information between and EHR (/PMS) and a diagnostic device where:
EHR Supports DICOM Modality Worklist and Integrates with a PACS
Based on IHE’s Unified Eyecare Workflow ‘real world model’ (RWM) I - Exchanging information between and EHR (/PMS) and a diagnostic device where:
EHR Implements HL7 Only (no DICOM support) and Integrates with a PACS)
Based on IHE’s Unified Eyecare Workflow ‘real world model’ (RWM) III - Sending a referral containing clinical information from and/or to any combination of the following practitioners: ophthalmologist, optometrist, general practitioner, specialist
- Sending a referral from a clinician to a healthcare service (eg - an ambulatory service centre when booking for cataract surgery.
- Exchanging information between and EHR and a clinical registry (eg - FRB!) - where:
The registry is highly structured;Not all registry required data points are routinely captured in a structured format in the EHR;Patient details must be de-identified. - Sending in bulk either all or part of an entire clinical record to a clinical registry.
- Sending select data elements from select patients (de-identified) in bulk to a research institute / life science body.
- The referral of a patient to a clinical trial.
- Automating prior authorization for anti-VEGF injections.
- Real time clinical trial recruitment using CDS hooks (eg for treatment-naive wet-AMD)
- SMART on FHIR app to compile and display relevant information into a single screen / application
外部链接
- Post title:Eyes on FHIR
- Post author:Kopei
- Create time:2022-11-20 00:00:00
- Post link:https://kopei.github.io/2022/11/19/Standard-2022-11-20-Eyes-on-FHIR/
- Copyright Notice:All articles in this blog are licensed under BY-NC-SA unless stating additionally.
Comments