Patent documents consist of well-structured descriptions of the technology innovation and the research results. Patent documents are typically written with specific terminologies which often require the expertise of domain experts to interpret and organize for patent analysis. The rapid growth of number of patents applications and registry increases the difficulty in extracting and synthesizing knowledge from patents. The aim of this paper is to construct a framework of expressing and extracting accurate knowledge from patent documents that are related to dental implant connections. First, domain related patents are collected from the United States Patent and Trademark Office. Text mining techniques are used to analyze the key words in the sample set of patents. An ontology is created to express the domain knowledge by using a new ontology engineering technique. In recent years, the aging population has increased the demand for new products and innovations in the dental implant industry. Dental implants are a global, medically accepted treatment used to restore the human masticatory function. The patents collected in the domain of implant connections are used to test the validity and reliability of the ontology development framework. The study of patents related to dental implant connections is considered a critical part of advancing dental implant devices since these designs and innovations are subjected to extensive clinical trials. Given an accurate ontology schema, dental researchers correlate successful prior art designs to successful clinical trials to better understand the interaction between a new artificial body part and the response of the surgical implantation. The value of a patent increases dramatically if the device can be safely used by dentists to treat patients with fewer failures and longer utility.