慢性腎臟病（Chronic Kidney Disease, CKD）的照護與日常飲食密切相關，須 嚴格限制食物種類、份量、與營養素攝取。CKD 飲食諮詢是多知識源及知識密集問 題，本研究以開發知識系統的觀點，整理「問題解決」所須的相關資料成為知識模型，另考量提升知識分享與再利用，因此採用本體方法並利用OWL（Web Ontology Language）做為塑模語言，開發為飲食諮詢知識系統。本研究初期建立的模型，已先排除合併症與併發症等干擾因素，再依據建置的知識庫開發為Web-based 知識系統， 協助使用者操作。在實驗評估部分，使用36筆個案資料進行測試，分別對臨床症狀、理論性的食物分類、敏感營養素限制值等進行推估，其次加入個案實際的飲食清單，檢驗理論值與實際攝取之差異。由實驗評估顯示，在排除前述干擾因素與部分非 醫學因素後，本研究已獲得初步成果，未來可再擴充問題解決的範圍，建置更完善的飲食諮詢系統。

The care for patients with chronic kidney disease (CKD) is closely related to the patient‘s daily diet management. The type, quantity, and nutrient content of the patient’s food intake need to be strictly controlled. CKD dietary consultation involves multiple knowledge sources and is thus a knowledge-intensive task. We approach the task from the perspective of knowledge system development by constructing related problem solving knowledge into a knowledge model. In addition, considering knowledge sharability and reuse, we adopt an ontological methodology and use OWL (Web Ontology Language) as the modeling language for the development of a dietary consultation knowledge system. In the early stage of the study, we have excluded the interference factors of complications & comorbidities from the knowledge model for scope definition. The constructed knowledge base is then developed into a Web-based knowledge system for better user experience. In terms of experimental evaluation, data from 36 case patients are used for testing. The system’s inference capability was tested, respectively, in clinical condition estimation, theoretical food group recommendation, and key nutrient restriction. The dietitian-advised diet lists are then added to examine the differences between the theoretical and actual intake values. The evaluation results show that, excluding the interference factors and certain non-medical reasons, the system has achieved the research goal of CKD dietary consultation. For future studies, the problem solving scope can be expanded to implement a more comprehensive dietary consultation system.