Design and Evaluation of the iMed Intelligent Medical Search Engine

Abstract

-Searching for medical information on the Web is popular and important. However, medical search has its own unique requirements that are poorly handled by existing medical Web search engines. This paper presents iMed, the first intelligent medical Web search engine that extensively uses medical knowledge and questionnaire to facilitate ordinary Internet users to search for medical information. iMed introduces and extends expert system technology into the search engine domain. It uses several key techniques to improve its usability and search result quality. First, since ordinary users often cannot clearly describe their situations due to lack of medical background, iMed uses a questionnaire-based query interface to guide searchers to provide the most important information about their situations. Second, iMed uses medical knowledge to automatically form multiple queries from a searcher' answers to the questions. Using these queries to perform search can significantly improve the quality of search results. Third, iMed structures all the search results into a multilevel hierarchy with explicitly marked medical meanings to facilitate searchers' viewing. Lastly, iMed suggests diversified, related medical phrases at each level of the search result hierarchy. These medical phrases are extracted from the MeSH ontology and can help searchers quickly digest search results and refine their inputs. We evaluated iMed under a wide range of medical scenarios. The results show that iMed is effective and efficient for medical search. All existing medical WSEs assume that searchers can form appropriate queries by themselves. However, most Internet users do not have much medical knowledge. Frequently, a medical information searcher has only a vague idea about the problem that he is facing and does not know the proper way to clearly describe his situation in sufficient detail. As a result, appropriate guidance is highly necessary during the medical search process. This can be illustrated by an analogy to the medical diagnosis process. During a doctor's office visit, the I. INTRODUCTION

Automatic Home Medical Product Recommendation

“…The input interface gathers this information by asking the user to make selections [3] and to input free text in three Web pages sequentially: a disease Web page, a symptom Web page, and an "other information" Web page. Often a HMP is used to treat a symptom (e.g., back pain) that can be caused by multiple diseases.…”

“…For the Web pages in C, we use the frequent term sequence method in [3] to drop noisy information (e.g., advertisements), perform standard pre-processing steps in Web information retrieval [3], and build an inverted index I i using the single-term vocabulary (i.e., the set of all distinct words). I i contains sentence IDs and its format is slightly different from that of traditional inverted indices [16].…”

“…Both our MedSearch system [19] and our iMed system [3,14] focus on disease diagnosis, and use disease diagnosis knowledge to help users find disease information related to their medical condition. The application in this work is recommending HMPs, which differs from disease diagnosis and is addressed with techniques different from those in [3,14,19]. [1] proposed the general concept of iPHR and suggested that HMP recommendation could be a useful function of iPHR.…”

“…Using medical knowledge, iPHR expands these topics into a more comprehensive set of topics. Next, iPHR will use a topic-selection interface [3] to obtain user input. In this way, ordinary people with little medical knowledge can avoid the traditional keyword interface that is difficult to use.…”

Triggers and Monitoring in Intelligent Personal Health Record

“…Our current iPHR system provides three functions: guided search for disease information [26], recommendation of home nursing activities [28], and recommendation of home medical products [30]. It follows the pull model of information distribution.…”