"When you hear hoofbeats, think of horses not zebras". Sometimes, however, physicians are confronted with rare diseases (the zebras). FindZebra is a tool for helping diagnosis of rare diseases. It uses freely available high quality curated information on rare diseases and open source information retrieval software (Apache Lucene Solr) tailored to the problem.
FindZebra is intended primarily for physicians and other professionals concerned with diagnosis of rare diseases. Users seeking information related to personal conditions are urged to consult with a qualified physician. FindZebra cannot be held responsible for any consequences following its use.
Orphan diseases affect more than 30 million Americans. 40% of patients are misdiagnosed or un-diagnosed. Traditional search engines need to be optimized to help diagnose rare diseases from symptoms. This is where FindZebra comes in! FindZebra is a symptom checker. Enter clinical and phenotype information, symptoms and findings and FindZebra returns potential diseases and genes involved.
Getting the most out of your search
FindZebra is a state-of-the-art diagnostic support tool featuring:
Symptoms filters. By selecting from the symptom list in the “Filters” menu, selected symptoms will be emphasised and search results refined.
Gene view. After an initial search one may choose "gene" in the "Group by" menu to get genes associated with the query. Also, a sorted list of relevant genes figuring in the top search results can be downloaded.
Keyword highlighting. The keywords most relevant to the query are highlighted in the text.
Flexible API. By using our API you can make custom queries that matches your specific requirements.
For example, you can take advantage of our advanced algorithms for gene and symptom extraction in order to quickly generate a list of known genes and symptoms associated with a given search string or disease.
Indexed medical sources
FindZebra.com contains more than 36,000 articles on approximately 7,000 diseases. Sources and approximate number of article from each source:
The Drug information has been obtained from Orphanet. Gene information has been obtained from Orphanet and OMIM.
Background information and citing this work
Radu Dragusin, Paula Petcu, Christina Lioma, Birger Larsen, Henrik L Jørgensen, Ingemar Cox, Lars K Hansen, Peter Ingwersen, and Ole Winther, FindZebra: a Search Engine for Rare Diseases, in International Journal of Medical Informatics, IJMI doi:10.1016/j.ijmedinf.2013.01.005 (2013). Journal linkPreprint arxiv
Follow up paper:
Radu Dragusin, Paula Petcu, Christina Lioma, Birger Larsen, Henrik L Jørgensen, Ingemar Cox, Lars K Hansen, Peter Ingwersen, and Ole Winther, Specialised tools are needed when searching the web for rare disease diagnoses, Rare Diseases, doi:10.4161/rdis.25001 (2013). Journal link
Svenstrup, Dan, Henrik L. Jørgensen, and Ole Winther. Rare disease diagnosis, A review of web search, social media and large-scale data-mining approaches, Rare Diseases 3.1 (2015): e1083145. Article link
Current FindZebra.com team
Dan Svenstrup (project management, research and development lead)
Ole Winther (project management, research)
Mads Emil Matthiesen (business management)
Henrik L. Jørgensen (team physician)
Thomas Nielsen (development)
Former FindZebra.com team members
Radu Dragusin, Paula Petcu, Christina Lioma, Róbert Kristiásson, Philip Pries Henningsen, Birger Larsen , Ingemar Cox, Lars Kai Hansen and Peter Ingwersen
The API is located at http://www.findzebra.com/api/call/[FORMAT]. The [FORMAT] specifies the return format and can be one of xml, json, csv, python, ruby and php. The api takes the following arguments:
q: The query.
fl: A list of document fields to return. Valid document fields are:
display_title: The document title.
content: The document text.
associated_gene: Genes associated with the disease.
symptom: known symptoms for the disease.
retrieved_date: the retrieval date for the document.
source_url: the original location for the document.
source: the document source (e.g. OMIM).
cui: the UMLS cui (if known).
score: the calculated relevancy of the document.
sort: the sorting field and order of the documents. If left out, the documents will be sorted by relevancy (score)
Enter a medical history in keyword form. Comma separate each distinct finding. Tip: After the initial search a Filter button appears. It can help you improve the search results.
What does the order of the search results mean?
FindZebra ranks results with most likely results in the top of the page. However, we recommend users to always assess all 20 results. Tip: Some search results have images that can help you quickly judge the match of a disease.
What search method should I use?
It is recommended to try more than one search. Combining different parts of the medical history allow you to focus in on certain dominant characteristics. Tip: Links the to article on the disease are listed under the disease name.
What can I do with the search results?
FindZebra is a tool to help diagnose rare diseases. FindZebra is intended as an aid for healthcare professionals and for patients who consult with their physician. Tip: Genes and market authorized pharmaceuticals, when available, are listed under the disease name.