Intended for healthcare professionals

Rapid response to:


How Web 2.0 is changing medicine

BMJ 2006; 333 doi: (Published 21 December 2006) Cite this as: BMJ 2006;333:1283

Rapid Response:

Ways to web 3.0

Dean Giustini describes in his excellent editorial the hype about the web 2.0 and possible implications of further developments of a „web 3.0“ in medicine. Convenient and open access to medical knowledge is the key to more efficient science and patient care. Web 2.0 / 3.0 suggest newer version of the internet but the evolution has been and will be continuous. The concepts of a Web 2.0 are enabled by easier tools to create and distribute own content for publication, such as wikis, social networks or RSS. They have emerged in the last three years. Contrary, the technologies and concepts of a semantic web (3.0) are already available. But the creation of semantics isn’t for free and it can’t be delegated nor to the machine nor to a non-expert.

Enrich textual knowledge

Information handling and knowledge management are boosted by the internet infrastructure, computing power, networking, and all mentioned approaches to enrich and structure the data. Semantic annotations, semantic links to related information sources, tagging and others transform the web towards a distributed database with defined entities and relations. But the reality isn’t a set of keywords, nor do we know all the keywords, topics or relations in advance. There is even evidence that “uncontrolled”, e.g. social tagging by an heterogeneous community is just as well as the use of an intelligent ontological framework handled by few. Both methods have to be used complementary. A controlled vocabulary such as SNOMED CT and domain ontologies are constantly developing. But in the end there will be some highly relevant information, up to now unknown and hidden data that is unmatched to the used models.

The search for information = knowledge creation

Finding best evidence in medicine isn’t easy. Turning data into information’s isn’t the strength of a machine (PC) but of human mankind, i.e. domain experts and the observations of the community (including physicians and patients). We definitely need the wisdom of the crowd and the experts. Semantic annotations will help to structure available evidence in order to build trusted clinical information.

In the next years we have to enable and motivate the actors to use semantic annotations to reach that long term goal. Unfortunately the benefits are not visible immediately. In the meantime the tools for data mining, search engine technology and the statistical methods to exploit semantically structured data will evolve. And Medicine will be a moving power for these fields of Applied Informatics.

Competing interests:
None declared

Competing interests: No competing interests

30 January 2008
Simon Hoelzer
CEO SwissDRG Inc.
3004 Berne