Olympus CEO Mark Gumz announced some exciting news today in a blog article/interview written by Steve Tobak.  Olympus is applying for FDA approval in digital pathology, and he also used digital pathology as an example to highlight “Enterprise Innovation” at Olympus.  Every FDA approval gained in digital pathology is a huge step forward for our industry and it’s great to see that Olympus is applying to the FDA for approval. The blog post, titled “Enterprise Innovation: A Key to Untapped Business Growth“ summarized a recent interview with the CEO.

Enterprise innovation is defined as “Transformative business processes, practices, organizational planning and models that enable a business of any size to operate more effectively, profitably, and/or competitively,” according to this article and a Harris Interactive survey of Fortune 1000 executives. Mark Gumz states in the article,

…The status quo is the greatest inhibitor to growth. Enterprise innovation is something that’s within our control as an industry and as a nation.

The FDA quote from Mark Gumz is as follows:

We’re currently applying for FDA approval for clinical digital pathology, the ability to send a clinical slide over the Internet. During a surgical procedure, a surgeon could take a biopsy and send it to a pathologist anywhere in the world to get a second opinion – do I take out more or not? That will ultimately be a huge cost saver. Ultimately, enterprise innovation has to come down to return on investment.

So what do you get if you put Mr. Gumz’s vision of “Enterprise Innovation” together with an application for FDA approval of clinical digital pathology?  My guess…not a whole lot in the near future!  There are too many unanswered questions and likely a lot of work to be done. Some of these unanswered questions are:

• What instrument, software, algorithm, technique, etc are being used in the application to the FDA? This is very confusing since Olympus America offers their own VS110 microscope-based slide scanner and software, distributes the NanoZoomer and it’s software from Hamamatsu, distributes image analysis software from Visiopharm on both hardware platforms, and still has server and viewer software from Bacus Laboratories, Inc (a company  they acquired in 2006).
• With whom are the clinical trials being conducted and what type of data is being gathered for application to the FDA? Several  competitors of Olympus have publicly announced and or have discussed details of their clinical trials.  For example, Aperio’s Project Pink is a primary diagnosis concordance study focused on breast tissue with four sites in the United States, and one of those sites is MD Anderson.  Omnyx announced in a recent press release the start of their clinical trials with three sites in the Unites States and one in Canada.  However, I have never seen or heard anything about clinical trials with Olympus.
• What is going to or has changed within the Olympus digital pathology product line to be more clinically oriented? They primarily sell hardware with a data management/viewer software solution that typically functions in a stand-alone research environment.  Clinically oriented hardware and software products from their competitors are very different and are designed with consideration for the histology lab and the diagnostic routine of a pathologist.
• What is being done to support laboratory workflow and interoperability within the pathology department and hospital? At Pathology Visions a Pathologist and I spoke with marketing staff at Olympus about this.  As of a few weeks ago Olympus did not have integration with any LIS vendors.  Also, they explained to the pathologist I was with that they do not support the DICOM format (WG-26 Supplement 145) recently approved for digital pathology.
• What about that ROI mentioned in the FDA quote above? I have not seen or heard of any marketed or published data showing Olympus’s efforts to create an ROI business model in digital pathology.  Maybe that’s in the works??

I realize that all companies working toward FDA approval are faced with many questions and tough regulatory hurdles, especially since the FDA has not provided any clear guidance to manufacturers on this topic.  However, based on the current public information out there about Olympus, their digital pathology product portfolio, and the information displayed on their website (which is 100% research and/or education focused) I have to imagine their FDA application is probably a rough draft and a long way off from being complete.

Pardon my skepticism… but another digital pathology middleware company?

On October 7, 2010 a new company, PixcellData, was formed out of Dublin.  PixcellData will be the third digital pathology middleware company out of Ireland, and will make the fifth middleware company within the digital pathology industry joining Apollo, AuroraMSC, SlidePath, and i-Path in this niche market space.

Middleware solutions are common in healthcare, and even in labs to help streamline workflow and improve interoperability and their is no doubt that this is needed.  Today, the marketed and perceived benefits of digital pathology middleware companies are:

• A universal viewer and/or data management system that supports all types of image files and formats
• Hosted and/or deployed IT enterprise solutions
• Scaleable, organized management of large data sets across multiple locations worldwide
• Faster delivery of data to clients than current hardware manufacturer solutions

These are nice benefits, however can they compete effectively against their middleware competitors and the ever growing marketplace of digital pathology hardware manufacturers?

Hardware manufacturers like Aperio, Bioimagene, CRi, Olympus,  Omnyx, Philips, etc would likely challenge that they offer all this and more within their solutions.  For example, Aperio’s SecondSlide has over 3000 users worldwide,  and Bioimagene’s PathXchange has over 8600 users worldwide collaborating in an easy to use environment that is scaleable, supports a variety of image files and formats,  and are designed to organize and deliver large datasets.  PathXchange even launched an “APP” this week- it seriously cannot get easier than that! On top of all of this, DICOM (a universal healthcare file format) was just been approved for whole slide images (WSI).  With a universal format established, WSI could soon be integrated in to radiology PACS, LIS, and other middleware systems improving the interoperability hurdle that our industry has been burdened with.   Now what are the benefits of a digital pathology middleware company?

Digital pathology middleware companies have achieved some business success through dissatisfied, frustrated current and prospective digital pathology customers.  It’s not that current manufacturers cannot do what middleware companies can do, it’s just that they often do not address the specific needs and fears of their customers well and often cannot deliver on all the “promises” made.  In this type of environment the middleware company could appear to be a knight in shining armor to the customer, the one who can answer all their specific needs and hold their hand throughout the process. Yet, this might only be a band-aid approach to a bigger problem.

The bigger problem is that the pathology community continues to be confused with the “who, what, where, when, why, and how” of digital pathology. WHO should we choose???   WHAT should we buy???? WHERE should we do this??? WHEN should we buy it????  WHY are we buying this????  HOW am I going to actually do this???? If these questions are not clearly answered and a plan established then often there is disappointment with the solution once an investment is made.   I do believe middleware has a place in laboratory workflow and interoperability; however, it’s a very tough niche market to define and an even harder one to compete in.

For over 10 years, the digital pathology industry has been riddled with vendors speaking their own language or languages (aka file format). Today a major milestone in interoperability was achieved with the final approval of Supplement 145; a DICOM universal format for whole slide images.  Although vendors will never give up their native tongue, this new supplement creates a universal language everyone can understand in digital pathology, even those outside of our industry.

WG-26, lead by Bruce Beckwith, MD, has worked hard for several years to create supplement 145.  Harry Solomon of GE Healthcare contributed a lot of help over the past year to move this to ballot and beyond.  Harry put forward the following comments in an email to WG-26 today:

I’d like to add my thanks to the members of DICOM WG-26 for the significant body of pathology domain knowledge that has been added to the DICOM Standard, both in this new Supplement 145, and in Supplement 122 in 2008.  That knowledge is now available to further improve interoperability in a field that was early to health informatics standards with the Systematized Nomenclature of Pathology, now known as the Systematized Nomenclature of Medicine (SNOMED), and is now poised for a significant leap with digital pathology.

Here are some statistics on that domain knowledge base:

• One major new DICOM Information Object Definition (Whole Slide Imaging) with new concepts to manage the huge WSI data set
• A major revision of a DICOM Information Entity (Specimen) with a rigorous information model
• 56 new data elements added to the DICOM Standard in 14 new or revised modules
• 7 new or revised DICOM structured data templates and 18 new defined value sets
• 80 new coded terminology concepts added to SNOMED, and 36 added to DICOM
• 40 pages of explanatory information about pathology workflow and imaging

Overall, this is another major step forward in the clinical use of digital pathology.  Thank you to everyone who worked hard to make this a reality!  The final document will be available within the next week.