Speech Recognition Sucks and It Costs More, Too!

Three articles in the December issue of the Journal of Digital Imaging discuss speech recognition, and only one is complementary. Let’s look at that one first.

Mika P. Koivikko, Tomi Kauppinen, and Juhani Ahovuo, all from the HUS Helsinki Medical Imaging Center of the Helsinki University Central Hospital, measured the effect of SR on report turnaround time (RTT). They found an 81% reduction in RTT, and found that “SR was easily adopted and well accepted by radiologists. . .with excellent end-user satisfaction.” The Helsinki team compared their old cassette-based transcription to SR using Philips SpeechMagic. They had some additional hurdles:

The Finnish language is challenging for SR because its vocabulary is exceptionally wide allowing many different words to evolve from one word body. HUS Helsinki Medical Imaging Center has actively participated in the development of a Finnish SR context for radiology.

The did improve RTT, as noted, although not uniformly:

During this study, the proportion of reports available within 1 h has rapidly risen from 26% (cassette Q1/2005) to 58% (SR Q1/2006). For nonurgent studies, such as most of our MR imaging procedures, the mean RTT still remained high (Table 2). In contrast, for typical high-priority worklists requiring online reporting (i.e., ICU or orthopedics), we measured an exceptional 53% reduction in RTTs and an increase from 34% to 65% in first-hour reporting. Thus, for our hospital, the increased number of reports available within 1 h from the completion of a study has proven a great improvement.

One has to wonder how much of the improvement came from eliminating the cassettes and going to a digital tank. The authors did appreciate the online editing inherant in SR.

The next article was not quite so kind to SR. Kimberly Voll, Ph. D., Stella Atkins, Ph. D., and Bruce Forster, M. D., from British Columbia, have more realistic observations. They see the potential, but realize the current inadequacies of SR:

The recent improvements of speech recognition (SR) technology have motivated the introduction of automated transcription software in lieu of human transcription. Speech recognition can offer improved patient care and resource management in the form of reduced report turnaround times, reduced staffing needs, and the efficient completion and distribution of reports. As the technology comes of age, however, with vendors claiming accuracy rates as high as 99%, the potential advantages of SR over traditional dictation methods are not being realized, leaving many radiologists frustrated with the technology.

The primary reason behind this apparent failure is accuracy. A 99%-accurate speech recognizer still averages one error out of every hundred words, with no guarantees as to the seriousness of such errors. Furthermore, actual accuracy rates in the reading room often fall short of 99%. Radiologists are instead forced to maintain their transcriptionists as correctionists, or to double as copy editors, painstakingly correcting each case, often for nonsensical or inconspicuous errors. Not only is this frustrating, but it is a poor use of time and resources. To compound matters, problems integrating with the radiology suite and the introduction of delays have further soured many radiologists on the technology. Those choosing to modernize their reading rooms with SR software are often plagued with difficulties, whereas those continuing to use traditional reporting methods have mixed incentives with respect to upgrading. Nonetheless, the potential benefits to radiology reporting from a hospital administration standpoint continue to motivate the adoption of SR technology. Thus, improving SR dictation is of particular importance.

Their solution:

As a partial solution to this problem, we have proposed a post-speech-recognition, statistical error-detection system for radiology. A previously unexplored area of research, this technique shows promise as an effective means to recover from the unacceptable accuracy rates of SR. By flagging potential errors, we can enhance the proofreading process, restoring the benefits of SR in resources saved. The result is a more efficient reading room and an improved experience with SR.

This solution dumps less on the radiologist, helping us in our new-found SR editing duties.

The piece-de-resistance, as far as I’m concerned, is found in the third article, by Pezzulo, et. al., from Brown University. The abstract says it all:

Continuous voice recognition dictation systems for radiology reporting provide a viable alternative to conventional transcription services with the promise of shorter report turnaround times and increased cost savings. While these benefits may be realized in academic institutions, it is unclear how voice recognition dictation impacts the private practice radiologist who is now faced with the additional task of transcription. In this article, we compare conventional transcription services with a commercially available voice recognition system with the following results: 1) Reports dictated with voice recognition took 50% longer to dictate despite being 24% shorter than those conventionally transcribed, 2) There were 5.1 errors per case, and 90% of all voice recognition dictations contained errors prior to report signoff while 10% of transcribed reports contained errors. 3). After signoff, 35% of VR reports still had errors. Additionally, cost savings using voice recognition systems in non academic settings may not be realized. Based on average radiologist and transcription salaries, the additional time spent dictating with voice recognition
costs an additional $6.10 per case or $76,250.00 yearly. The opportunity costs may be higher. Informally surveyed, all radiologists expressed dissatisfaction with voice recognition with feelings of frustration, and increased fatigue. In summary, in non-academic settings, utilizing radiologists as transcriptionists results in more error ridden radiology reports and increased costs compared with conventional transcription services.

Read that summary one more time! SR reports are costlier and have more mistakes! So the only benefit might be from increased RTT, but with an adequate transcription pool, even that falls by the wayside. For what it’s worth, the study used Agfa TalkTechnology version 2.1.28.

The article made a very profound discovery: “Our results suggest that radiologists are not good transcriptionists.” Duh.

Once again, when you look below the surface, SR just isn’t a good idea. It may improve turnaround time, but generally only when RTT was dismal to begin with. It just doesn’t work well, and in the private-practice setting at least, it is costly and error-prone. Of course, the hospital bean-counters don’t care about the added cost of a radiologist-transcribed report because….they don’t pay the radiologist! But outside of that, SR just doesn’t make sense. Not yet, anyway.

AMICAS To Buy Emageon!

Wonders in the PACS business never cease. I’ve been reporting on the merger/acquisition debacle concerning Emageon and HSS, which was scuttled by the imminent demise of Stanford Financial. But like the Phoenix, coincidentally, the code-name for AMICAS’ revolutionary Version 6 software, Emageon will rise from the ashes to be purchased by…AMICAS!

I just received an e-mail from Dr. Stephen Kahane, President of AMICAS (I’m sure I wasn’t the only recipient) which included the following paragraphs:

Emageon is an organization with a mission and goals that are very similar to AMICAS’. While AMICAS and Emageon have approached the market with different strategies, we believe that our strategies and our businesses are highly complementary.

Together, we believe that AMICAS and Emageon will be the undisputed independent leader in image and information management with over 1,000 customers. As one organization, we’ll bring together the best employees, customers, and solutions in image and information management – including radiology PACS, radiology information systems, cardiology PACS, cardiovascular information systems, referring physician tools, business intelligence tools, enterprise content management systems, and revenue cycle management systems.

From the Frequently Asked Questions:

    • Our combined business will have over 1,000 customers!
    • Our combined business will span the entire continuum of imaging customers – from large radiology practices, like Radiology Limited (Tucson, AZ), all the way to prestigious institutions such as The Johns Hopkins Hospital (Baltimore, MD).
    • Our combined business will have a comprehensive portfolio of image and information management solutions for both radiology and cardiology.


We‘ll move quickly to bring Emageon and AMICAS together. There won’t be an “Emageon” division of AMICAS. There will be ONE AMICAS: one team, one plan, one mission. This acquisition is a fairly significant acquisition for a business of our size, so it will take time and a lot of hard work. We expect the integration to be substantially complete in the weeks following the closing and fully complete within 180 days.

What about product overlap?

There is remarkably little product overlap between the companies. The core products at AMICAS include radiology PACS, radiology information systems, revenue cycle management solutions, referring physician solutions, and business intelligence solutions. The core products at Emageon include radiology PACS, cardiology PACS, cardiovascular information systems, and enterprise content management systems. The only product overlap is with radiology PACS – the other solutions are already highly complementary to one another.

The radiology PACS solutions from both AMICAS and Emageon will continue to be sold, deployed, supported, and maintained – both solutions are robust offerings with very happy customers!

Given the slaughter of Emageon’s stock price, it was a bargain waiting to be plucked, and AMICAS apparently had the cash on hand to take advantage.

Here’s my personal assessment, neither reviewed or endorsed by AMICAS: This move bought AMICAS a significant increase in its customer base. That alone was probably worth the expenditure. AMICAS now (finally) will own its own cardiac package. You might recall that Emageon bought Camtronics, and their cardiac software and EMR, in November, 2005, for $40.4 million.

As for PACS, I would expect AMICAS to support the Emageon product as they promised, but. . . I haven’t played with Emageon PACS in quite a while. The last I heard about it was from Dr. Elliot Fishman during a Johns Hopkins lecture in 2007. He wasn’t terribly complementary, saying that it choked if they tried to load anything larger than a 100-slice CT, forcing him to use the Siemens InSpace/Leonardo station. Now, Elliot loves InSpace, so maybe there was a little hyperbole involved. Still, I think we can probably label Emageon PACS as “End of Life”. Knowing the principals (and the principles) of AMICAS, I expect them to keep Emageon PACS sites up and running for several years, as they have promised. Ultimately, I would expect those sites to be moved to AMICAS PACS (Version 6), although I have no idea how much they will or won’t charge for this privilege. Again, this is only my personal speculation, with absolutely no confirmation from AMICAS whatsoever.

I have heard that the Emageon “back end” may have some advantages, although I don’t know much about that at this time.

To me, the most important thing to come out of this deal is an unsaid message. Amicas loses deals here and there because they are “too small” or “likely to be taken over”. With the Emageon purchase, they have made a loud statement: They are solvent, they are stable, and they are the big company buying the smaller one, the pursuer and not the prey, so to speak. At worst, they got Camtronics for less than what Emageon paid for it in 2005, and they significantly increased their customer base.

I can promise, though, that some folks made some money today. Emageon stock (EMAG) went from $0.80 to $1.68 at 1:10 PM today, right around the time of the press release. I wasn’t one of them, by the way. The stated purchase price is $1.82 per share, for a total of about $39 million. I wonder if AMICAS could have negotiated it down a bit further.

Any chance of AMICAS headquarters moving down to Birmingham? The weather is so much better there than in Boston. . .

The PACS of Your Life (Good Riddance?)

Another breaking point;
An upgrade redesigned.

Techs grab you by the wrist;
And ask why you’re behind.

So do the best beta test
And don’t dare whine.

You mustn’t question
How the vendor wastes your time.

It’s never quite predictable
Just when the PACS will die.
You know it’s going to mess with your mind.

So take the cine loops
And CT’s in your mind.

Hanging protocols
Might work some other time.

Worklists that never run
And buttons galore;

For all they’re worth,
No one knows what they’re for.

It’s never quite predictable
Just when the PACS will die.
You know it’s going to mess with your mind.

(music break)

It’s never quite predictable
Just when the PACS will die.
You know it’s going to mess with your mind.

It’s never quite predictable
Just when the PACS will die.
You know it’s going to mess with your mind.

This Just In: Stanford Charged With ‘Massive Fraud’

Thanks to an anonymous tipster, who sent me the link to this piece from pMSNBC:

Federal regulators on Tuesday charged Texas financier R. Allen Stanford and three of his companies with a “massive” fraud that centered around high-interest-rate certificate of deposits.

In a complaint filed in federal court in Dallas, the Securities and Exchange Commission alleged Stanford orchestrated a fraudulent investment scheme centered on an $8 billion CD program that promised “improbable and unsubstantiated high interest rates.”

Stanford’s assets, along with those of the three companies, were frozen. Stanford’s companies include Antigua-based Stanford International Bank and broker-dealer Stanford Group Co. and investment adviser Stanford Capital Management, which are both based in Houston.

Looks like Emageon dodged a bullet. I still wonder if the $9 Million check from HSS cleared….

Emageon’s Suitor’s Financier’s Tangled Troubles

Sir R. Allen Stanford, Stanford Financial Group, probably waving, not thumbing his nose at Emageon.

Photo courtesy of http://online.wsj.com

Three Wall Street Journal articles have appeared in the past two days, outlining a probe into the Stanford Financial Group, which owns the bank that reneged on financing HSS’s purchase of Emageon.

The story from Friday, February 13, notes that Stanford is being examined by the SEC, the Financial Industry Regulatory Authority (FINRA), and the Florida Office of Financial Regulation.

The focus of the probe isn’t clear. Mr. Stanford’s firm offers many financial services, chief among them wealth-management services, and markets certificates of deposit through his offshore bank, Stanford International Bank Ltd., in Antigua. The CDs offer unusually high and consistent returns. Mr. Stanford controls both the bank and the financial group that markets the CDs the bank issues.

By Valentine’s Day, yesterday, WSJ reported that the FBI had entered the probe. It seems that Stanford Financial Group had recently been telling its depositors that they couldn’t redeem their CD’s for two months. Sir Allen emailed his customers, telling them that he had been “infusing cash” into the Antigua-based Stanford International Bank, Ltd. How did Stanford reach the FBI’s radar? It seems that like our old pal Bernie Madoff, some of Stanford’s returns are too good to be true given the investments he claims to have made:

Stanford Financial Group previously has been in the sights of various regulatory and law-enforcement agencies, according to the people familiar with the matter. Those people say the agencies now are focusing on certificates of deposit, which are marketed by the financial group’s wealth-management arm and sold by Mr. Stanford’s Antiguan bank. The CDs offer unusually high returns; for example, as of Nov. 28, a one-year, $100,000 CD paid 4.5%.

“The first thing that grabs your eye is the business model,” says Alex Dalmady, an analyst who unveiled concerns about Stanford International Bank in the magazine VenEconomy Monthly but isn’t involved in the investigation. “Taking deposits and playing the stock market — this is way too risky. “

Stanford’s Web site says that the bank has invested in a diversified portfolio including stocks, while banks generally make money by lending. A memo on Stanford’s Web site says that the bank has never made a structured loan or a commercial loan. . .

“We want our depositors to know that SIBL had no direct or indirect exposure to any of Madoff’s investments…Also, SIBL has never made a structured loan or a commercial loan. All loans are cash secured and to SIBL clients only at a maximum 80% loan to 100% cash collateral ratio.”

Now here is where it gets interesting. A second WSJ article from February 14, 2009, tells us:

An offshore bank at the center of two U.S. federal investigations recently curtailed financing commitments to two small American companies, regulatory filings show.

Stanford International Bank Ltd. of Antigua recently failed to provide some $16 million in funding to a small Florida telecommunications firm, while a small Alabama health-care firm said it was unable to complete a roughly $62 million merger after funding fell through. Stanford International had previously planned to provide funding to complete the deal, according to the health-care firm.

It was understood that Stanford was funding Health Systems Solutions’ bid for Emageon. But the true relationship appears to be a little deeper:

Stanford also owns a majority of the shares of Health Systems Solutions Inc. of New York City, which is traded on the OTC Bulletin Board. In October the firm agreed to acquire Emageon Inc. of Birmingham, Ala, but the deal was terminated Friday with Emageon attributing the development to Health Systems’ inability to obtain funding on or before the closing deadline of Feb. 11. Executives of Emageon couldn’t be reached for comment and Health Systems executives didn’t respond to requests for comment.

Hmmmm. Interesting. As an aside, Stanford had a similar relationship, and a similar deal-retraction with Elandia International, a communications firm:

The Florida firm losing $16 million in financing from Stanford International is called Elandia International Inc. of Coral Gables, which trades over-the-counter on the so-called pink sheets. Elandia says it controls a collection of small telecommunications firms in Latin America and the South Pacific. Regulatory filings show that the Elandia’s chief financial officer is James M. Davis, who is also chief financial officer of both Stanford International Bank and Stanford Financial Group.

Efforts to contact Elandia executives by phone and e-mail were unsuccessful.

In a new SEC filing, Elandia also said Stanford International Bank had also agreed to convert an outstanding $12 million loan it had made to Elandia into shares of Elandia equity. Such debt-equity swaps often take place when borrowers lack the cash to pay loans back.

Since Emageon got their $9 Million payoff for the deal’s collapse (I assume the check cleared), they came out of this “merger” better than they might have had it actually gone through. I have just two words of advice, though: due diligence….. Someone should have had a deeper look into the inner workings of HSS and Stanford. But I guess it’s rather rare for people about to get married to look deeply into anything but the eyes of their beloved.