An Ontology-Based Approach for the Development of Residents Assessment Knowledge for Long-Term Care Institutions

[[abstract]]近幾年來，全球人口有逐漸高齡老化的趨勢，衍生出許多健康照護的問題，隨之而來的民眾逐漸意識到老人的醫療照護問題極為重要。由於在台灣人口與家庭結構變化的關係，機構式的長期照護服務模式已成為是不可缺少的方式。長期照護機構中最重要的就是評估住民的身心狀態和照護住民的日常生活，因此本研究目的是發展以本體論方法於長期照護中機構式的住民評估的靜態與動態知識。 本研究主要是提出三階段的本體發展流程，能與不同領域的專家做溝通。資料收集透過訪談長期照護機構取得靜、動態知識和文獻收集的方式，靜態知識包括住民管理、評估量表、照護計畫與照護週計畫表，其中評估量表是參考「日本介護評估量表」架構中的五大構面「身體機能與起居動作、生活機能、認知機能、精神行為障礙、對社會及生活的適應」，及「高齡者照護計畫分類分級量表(TAI)」的分類分級概念，規劃成符合台灣長期照護機構的評估量表，並了解評估結果的分數、等級和類型所代表的照護需求為何，進而產生屬於住民個人的照護計畫與照護週計畫表。而動態知識則以SWRL推論機制來產生新知識，最後塑模為長期照護機構住民評估知識模型。 以本體的方法論建置長期照護住民評估的知識模型，本體論具有明確定義領域知識的相關概念、關聯、屬性、規則的功能，能清楚的描述每個領域的知識內容，並可將此領域的知識與他人分享，因此本研究是以本體論之方法把知識進行塑模，另外使用實際例子來推論出新知識，主要可以提供予照護人員與系統開發人員溝通與了解此專業領域的知識，並共同分享此本體模型，使長期照護領域的知識能達到知識分享與應用的價值，讓住民得到最適切的照護。 The problem of global aging population derived many health care problems. Most countries aware of the healthcare for elderly population are an extremely important issue. Due to the changed of the population and family structure in Taiwan, long-term care institution has become an indispensable service in current community. The most important tasks for long-term care services in the institution are to assess the physical and metal states, and provide the suitable cares for residents. Thus, the purpose of this study is to develop static and procedural knowledge with ontology technique for the assessment of institutional residents. First, we proposed a three-stage ontology development approach that allows members of the interdisciplinary teams to elicit static and dynamic knowledge semantically. Next, the data collection process involves obtaining static and dynamic knowledge, as well as conducting a literature review, the static knowledge includes resident management, assessment scales, care plan and care weekly plan. Herein, the assessment scales that we refer in this study are Japanese Nursing Care Assessment and Typology of the Aged with Illustrations (TAI). Japanese Nursing Care Assessment provides the assessment concept of five dimensions including physical function, daily actions of life function, cognitive function, mental and behavioral disorder, social and life adaptation. TAI provides the concept of classification and grading for residents. We design the assessment scale suitable for Taiwanese long-term care institutions to understand what caring requirements are indicated by the score, type, and grade of the evaluation. Third, we modeled dynamic knowledge as SWRL rules for the description of resident assessment procedure. In this study, the data ontology comprises relevant concepts, associations, attributes, and rules in clearly defined areas of knowledge. In addition, we examine real cases to obtain new knowledge. This can enable care personnel and system development personnel to communicate and understand professional knowledge and to share the model. Furthermore, the model demonstrates the value of sharing and applying knowledge in the field of long-term care, ultimately ensuring that residents of long-term care institutions receive the optimal care