An international dermatological image atlas on the WWW: practical use for undergraduate and continuing medical education, patient education and epidemiological research.

We describe the development of an image database DOIA (Dermatological OnlIne Atlas) and present several spin-off projects using images of the atlas, e.g. student education using the atlas including results of an questionnaire evaluating computer-literacy, prerequisites and interests of students for using computers and the World-Wide-Web (WWW), a patient information system and an experiment to collect epidemiological data from patients with dermatological diseases via WWW. The database, available on the WWW at http:@www.derma.med.uni-erlangen.de, contains about 3,000 clinical images covering more than 540 dermatological diagnoses. It is designed for worldwide use; international submissions are encouraged. One aim of the project is to compile an international reference for dermatological images, containing images of high educational quality and also covering conditions on different skin types and rare diagnoses which are not commonly illustrated in ordinary textbooks. All images were originally mapped to the Erlanger Diagnosis Code, which is a proprietary modified ICD-9 key, later also to the UMLS (Unified Medical Language System). In addition, images are described with keys for the location, physical attributes of the location and clinical and histopathological features of the lesion. In order to facilitate the integration of the atlas into other web-based medical resources and to allow easy access to additional information, the Erlanger Diagnosis Code was mapped to the CUIs (unique concept identifiers) of the UMLS Metathesaurus. One purpose of the UMLS is to allow conversion of terms from one controlled medical vocabulary to another, thus, mapping of our diagnosis code to the UMLS CUIs allows simultaneous search for a given diagnosis in a number of other databases and also access to our image database from other databases. Mapping was successful for 619 out of 1383 dermatological diagnosis terms. For images with these diagnoses we are able to provide a hyperlink to other databases available on the Internet, such as MEDLINE, PDQ and OMIM, with automatic retrieval using the preferred vocabulary of the respective database. By grouping all diagnoses into sets with similar morphologies we further integrated a differential diagnosis mode. In order to educate patients via the Internet, a separate patient information system has been developed, using images of the electronic atlas. As an experiment to explore the feasibility of the Internet to gather epidemiological data from patients, users are asked to complete an electronic questionnaire covering signs for atopy. We conclude that an online image atlas has multiple educational, clinical and research applications.