The L&C Newsletter is issued bi-monthly and gives information on our company and our solutions for intelligent information enrichment and retrieval in the medical and pharmaceutical sector. We will also give updates on trends and technological innovations.

In case you want to subscribe to the newsletter, please send an e-mail to: This e-mail address is being protected from spam bots, you need JavaScript enabled to view it

L&C joins HIMSS Taskforce on BioTerrorism

Dr. James R. Flanagan, MD PhD, CMO at L&C, has been invited to join the HIMSS Task Force that will develop activities to help the healthcare industry create and adopt a national health information infrastructure (NHII) in the US.

"The NHII task force is yet another example of HIMSS' continuing efforts to address important and timely issues affecting the nation's healthcare industry," said H. Stephen Lieber, HIMSS president and CEO. "The leaders on this task force represent the best minds in the industry on issues such as technology, standards, applications, systems, and laws supporting all facets of healthcare."

EMR vendors are looking to NLP for Increased User Adoption of the EMR

The adoption of the EMR has been rather slowed by the change in culture and habits required by widely available technology. Many articles suggest that the solution to EMR adoption problems requires changing culture and habits. However it is the "structured input" nature (e.g., "pick lists") of the technology that demands changes in the habits of clinicians and creates the enormous barrier to EMR user adoption. Structured input is rather time-consuming and often prevents the physician from recording the actual information since it allows only the options available in the system. This can result in both incomplete and incorrect information. For example, it is quite difficult to capture information typically found in an HPI (history of present illness) without using free text to tell the story.

To overcome these problems EMR vendors are looking for ways to allow physicians to work as much as possible the way they are used to. In many cases this means allowing natural language expression in the EMR. Ideally, this must be coupled with methods to extract information from the natural language input. This is where Natural Language Processing technology comes into play. Advanced NLP technology allows physicians to work the way they find most efficient while still giving the EMR the possibility to use this information in an intelligent way (e.g. for billing purposes, for clinical decision support, for quality control, for research, etc.). NLP-powered EMRs also allow input from voice recognition applications although the NLP then depends on the quality of the voice recognition software.

The EMR will not enjoy a widespread user adoption until this kind of technology provides a natural interface for physicians. L&C's advanced NLP technology has already been selected by some of the largest EMR vendors worldwide to this end.

The benefits of NLP-powered search engines

Traditional search engines use keyword or statistic based techniques to match documents to user queries. The performance of these search engines is very low due to a low recall (they do not retrieve all relevant documents) and a low precision (they return irrelevant documents, also called 'false hits').

To make this clearer, we can take the example of a physician who is looking for information on infectious heart diseases. With conventional search methods, the information system will provide documents containing the words "infection" and "heart." This would include documents containing phrases such as "the antibiotic got right to the heart of the infection," but would miss documents containing "bacterial endocarditis."

Advanced Natural Language Processing (NLP) technology makes it possible to solve these problems. It creates a reasoning environment that gives computers an understanding of a specific domain (e.g. medicine, legal, finance, etc.) so computers can use this to automatically process natural language. The result in the example above is that the search engine will know that 'bacterial endocarditis' is in fact an infectious heart disease, but a document with "the antibiotic got right to the heart of the infection" does not deal with infectious heart diseases. Both recall and precision of NLP-powered search engines will be much higher if compared with traditional search engines. This leads to more complete information, no waste of time browsing through irrelevant documents, faster access to relevant information and ultimately a huge increase in business performance.

Coding for billing using NLP
: LinkCode

One example we can highlight is the use of natural language processing (NLP) to extract facts for business processes such as coding for billing. Without NLP, the process of coding for billing requires enormous numbers of man-hours reading through text records of patient encounters on the part of clerical staff trained in both medical terminology and in rules for billing. A better solution will involve the coupling of two advanced technologies. One is the use of NLP to extract key data from text. This technology uses semantic networks called Ontologies and statistical tools for disambiguation to overcome the problem of varied expression contained in text. Once the key facts are extracted from the text they can be coded using a variety of standard controlled vocabularies and fed to a rules engine that computes billing rules. Together, these two technologies can replace the enormous manual effort now expended in this task. This will result in huge savings, more reliable information, and a more complete picture of all patient encounters.