NLP and MedlEE
January 12, 2011
NLP International Corp. and its MedlEE product have popped on and off our radar several times in the last two years. NLP offers natural language processing technology and service tailored to the needs of the health care market. MedlEE has roots that reach back to Columbia University. The firm:
With its unique deployment model NLP International makes this world class solution available for through our MedlEE Portal. The MedlEE Portal is a SaaS offering the has applications ranging from quality to semantic search and retrieval to computer assisted coding and meaningful use… The MedlEE Natural Language Processing engine codifies standard text documents for data extraction, thereby enabling Discrete Reportable Transcription (DRT). MedlEE was developed over the course of 20 years by Columbia University in New York and is a powerful patented NLP engine that automates analytics, reporting and alerting for various reports for billing and for meeting various requirements.
There appears to be interest in the use of NLP and semantic technology in various health care applications. StenTel Corp., one of NLP’s MedlEE partners, says:
MedlEE NLP turns unstructured, dictated medical narratives into easily retrievable accurate data to support multiple health care systems in the hospital to enhance patient safety, quality assurance, diagnosis and prognosis support, billing and reimbursement administration. The physician is not required to change work habits.
What we found interesting is that NLP International has gone “silent”. Is an acquisition or other deal in the works? Is there another reason for the low profile?
We will monitor the situation.
Stephen E Arnold, January 12, 2011
Freebie
Comments
4 Responses to “NLP and MedlEE”
[…] This post was mentioned on Twitter by Stephen E Arnold, Efrosyni Adamides. Efrosyni Adamides said: NLP and MedlEE : Beyond Search http://bit.ly/g90H6O […]
NLP in health care application!? Interesting, but why NLP International is quiet?
Stumbled across this one today, and thought I’d provide an update. MedLEE is a best of breed Natural Language Processing technology developed out of Columbia University as you’ve stated. The technology is able to ingest unstructured clinical notes data, understand it, and associate the applicable medical codes to the text.
In theory, applying the applicable codes to the clinical notes data, should make it exponentially more useful in terms of an organizations ability to mine it, but it also creates a problem that has been addressed on more than one occasion on this blog.That problem is Scalability. An unprocessed medical record is relatively small, but, when processed by MedLEE, the size of that record may be more than quadrupled.
Despite the fact that the record is in theory exponentially more usable, it’s also exponentially harder to deal with in terms of the hardware and computational power required to mine the data in a meaningful way.
The answer? Perfect Search Corporation. Perfect Search is a technology that has likewise been featured on this blog and offers some unique attributes in terms of their software based innovation. Namely, the ability to get to data at least ten times faster, be much more comprehensive in terms of the content searched, and operate on literally 90% less hardware than traditional “search” solutions.
Through a partnership between Perfect Search Corporation and NLP International, the organizations have been able to take MedLEE to market and have strategic implementations/partnerships with some of the largest medical transcription services in the United States and are currently reviewing strategic partnership opportunities with other’s in the Healthcare space.
This is the solution demo’ed for the Arnold IT team at the Perfect Search Offices in Utah a couple of months ago, and one that continues to draw praise from those utilizing it today.
The Perfect Search NLPI solution provides a data mining and analytics tool that allows clinicians, research physicians, hospital administrators, and others the ability to construct complex, ad hoc queries, against databases of literally millions of MedLEE processed clinical notes, and get back search results in sub second time.
I don’t mean to get into a heavy handed sales pitch, but wanted to provide an update on where the technology/company is at, and would offer that any one interested in a demo, is free to contact me directly.
Was nice to see you folks in Utah the month before last, and I’ll keep you updated on our progress.
NLP International lost their license MedLEE due to contractual breech with Columbia