Searching for medical information on the Web is a challenging task for ordinary Internet users. Often, users are uncertain about their exact medical situations, are unfamiliar with medical terminology, and hence have difficulty in coming up with the right search keywords. An intelligent medical search engine is specifically designed to address this challenge. It uses several techniques to improve its usability and search result quality. First, it uses an interactive questionnaire-based query interface to guide users to provide the most important information about their situations. Users perform search by selecting symptoms and answering questions rather than by typing keyword queries. Second, it uses medical knowledge (e.g., diagnostic decision trees) to automatically form multiple queries from a user' answers to the questions. These queries are used to perform search simultaneously. Third, it provides various kinds of help functions. For instance, if a user has made incorrect selections, does not know which symptoms to select, or cannot answer a question, he or she can invoke the search engine's help functionality and describe his or her health situation there. Then he or she will see multiple lists of help information, e.g., one list for relevance of unselected symptoms, and another list for relevance of alternate responses (to his or her selections). The contents of those lists are ranked using medical knowledge and specially designed medical information retrieval techniques.
Gang Luo and Chunqiang Tang (2008). "On Iterative Intelligent Medical Search". Published in Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2008.