PhD student Health Data Science, Faculdade de Medicina da Universidade do Porto, Portugal.
Associate Professor, Faculdade de Medicina da Universidade do Porto, Portugal.
Stud Health Technol Inform. 2024 Aug 22;316:230-234. doi: 10.3233/SHTI240387.
One approach to enriching the Learning Health System (LHS) is leveraging vital signs and data from wearable technologies. Blood oxygen, heart rate, respiration rates, and other data collected by wearables (like sleep and exercise patterns) can be used to monitor and predict health conditions. This data is already being collected and could be used to improve healthcare in several ways. Our approach will be health data interoperability with HL7 FHIR (for data exchange between different systems), openEHR (to store researchable data separated from software but connected to ontologies, external terminologies and code sets) and maintain the semantics of data. OpenEHR is a standard that has an important role in modelling processes and clinical decisions. The six pillars of Lifestyle Medicine can be a first attempt to change how patients see their daily decisions, affecting the mid to long-term evolution of their health. Our objective is to develop the first stage of the LHS based on a co-produced personal health recording (CoPHR) built on top of a local LLM that interoperates health data through HL7 FHIR, openEHR, OHDSI and terminologies that can ingest external evidence and produces clinical and personal decision support and, when combined with many other patients, can produce or confirm evidence.
丰富学习型健康系统（LHS）的一种方法是利用可穿戴技术的生命体征和数据。可穿戴设备（如睡眠和运动模式）收集的血氧、心率、呼吸频率和其他数据可用于监测和预测健康状况。这些数据已经在被收集，并且可以通过多种方式用于改善医疗保健。我们的方法将是健康数据的互操作性，与 HL7 FHIR（用于不同系统之间的数据交换）、openEHR（用于存储与软件分离但与本体、外部术语和代码集相关联的可研究数据），并保持数据的语义。openEHR 是在建模过程和临床决策中具有重要作用的标准。生活方式医学的六个支柱可以作为改变患者如何看待其日常决策的首次尝试，影响其健康的中长期发展。我们的目标是基于 CoPHR 开发 LHS 的第一阶段，该 CoPHR 建立在本地 LLM 之上，通过 HL7 FHIR、openEHR、OHDSI 和术语进行互操作，这些术语可以摄取外部证据并生成临床和个人决策支持，并且与许多其他患者结合使用时，可以生成或确认证据。
