Topics:

First-Ever Mammography Exchange Will Make Prior Exams Available

First-Ever Mammography Exchange Will Make Prior Exams Available

Women who come to Mercy Hospital Medical Center in Chicago for mammograms usually aren’t repeat patients. If they lack medical insurance and are lower-income, these patients tend to get such care wherever and whenever they can — from free clinics to one-day events in the basement of the local church.

That means their mammogram images are stored all over the place, which makes gathering past testing for comparative purposes — essential to a full and proper exam — challenging. The process can take weeks, if images can be gathered at all, which compromises mammographers’ ability to make an efficient and accurate diagnosis.

Eileen Knightly, director of breast center and women’s health at Mercy Hospital, has seen many cases where patients actually forget the site of their mammograms. That’s why she is among many who are excited about the proposed pilot of the Chicago Mammography Image Exchange (CMIX), the first shared image superhighway of its kind in the nation. The program, starting in May, will leverage the energy and expertise from the Chicago Health Information exchange, in which 80 Chicagoland hospitals have agreed to participate. The proposed pilot will involve nine sites in Chicago.

Reducing Healthcare Disparities

Those participating in the exchange — which will store mammogram images in an online “cloud” accessible from any participating breast center — believe their work will have immeasurable benefits, cutting costs, streamlining processes and improving clinical outcomes. Most importantly, it may address the issue of healthcare disparities such as the 62 percent greater likelihood that black women will die of breast cancer than white women in Chicago, according to the Sinai Urban Health Institute's 2006 report, "Breast Cancer in Chicago: Eliminating Disparities and Improving Mammography Quality.”

The idea for the exchange was hatched at a meeting I had with the chief medical officer of a large academic medical center in Chicago. He told me the mammography problem was severe, and that institutions like his needed interoperability to solve this issue of prior images being available at the point of care. At that moment, we both jumped out of our seats and said, “Hey, we can do this!”

I have 15 years of experience practicing medicine in some of Chicago’s most under-resourced neighborhoods and expertise in image exchange from working at Merge. And a significant part of my role has involved figuring out what problems the technology’s interoperability could solve, with a priority on meaningful applications useful to women across the socioeconomic spectrum.

These possibilities have been further underscored through my role as board chair for the Illinois Health Information Exchange, a statewide, secure electronic network for sharing clinical and administrative data among health care providers.

A Cumbersome Process

Merge Healthcare brought the concept of an image exchange to founding members of the Chicago Breast Cancer Task Force and demonstrated the possibilities to the group of stakeholders who are now helping the idea move forward.

An image exchange will allow hospitals access to imaging that’s occurred for patients over their lives, and will allow the mammographer to make more accurate and more efficient diagnoses. Women without access to healthcare get care wherever they can — from one facility, and then another facility, and then come to settings like Mercy, Knightly said. Hospitals need to compare that mammogram over time to make the best diagnosis. Chasing those images is very inefficient and often takes weeks. Too often, the woman waits for an understanding of what needs to happen next.

The current workflow at breast centers is extremely inconvenient and costly. One center faxes a release to another, and images then are hand-pulled, put on a CD and sent through the mail and downloaded onto the requesting center’s PACS.

The current system is cumbersome, with lots of potential for error. Things get lost in the mail, or an image is misfiled somewhere — and then hospitals can’t find it. Knightly said she can cite multiple cases in Chicago where flooding has occurred and film gets destroyed. Even if images are found, they’re not usable. If they were in a cloud such as the CMIX, there’s no hurting these images. It would certainly make for more efficient care, and would improve processes in breast centers dramatically.

When it takes more than 30 days to find images, best practices dictate that a repeat screening or other diagnostic procedures be performed. This further delays access to care and reinforces healthcare disparities — in addition to creating unnecessary costs. A lot of women get additional imaging, and they may not even need it. That wastes resources that are already very limited.

Sometimes, after those weeks of waiting, the patient cannot even be found. According to Knightly, if a hospital does a mammogram today and doesn’t get back to the patient for 60 or 90 days, the odds increase that the patient cannot be found. The goal of the exchange is to make sure that women are taken care of in a higher-quality manner.

The Potential for Streamlining

When patients can’t remember where they’ve had mammograms done in the past, the disparate radiology storage systems linked to the exchange will be able to “talk” to one another, find the patient’s records and pull them from the vendor-neutral archive where a copy of the mammogram will be stored.

If providers can query the cloud for a patient’s images, CMIX will be a great success. The whole concept is really exciting, because a lot of worlds are being brought together in a collaborative approach to help improve the quality of healthcare.
 

Cheryl Whitaker, MD, MPH, FACP, is chief medical officer of Merge Healthcare.

Disclosures

Quite interesting and about time since digital mammography came into exstence. The only problem, I can see, is that many women who goto different facilities are having their exams performed on different vendors equipment. Each vendor has made dedicated and enhanced algorithms for viewing their images. While DICOM is standard, these companies have their own functionality on their own equipment and a generic workstation, although excellent, does not have the capabilites to view each imaging set to the best ability.
Alan Melton (not verified) @
Thank you for your comments. Great strides have been made with exchanging DICOM. We now have more vendors with published DICOM conformance statements and better support for standards that support exchange. We now have more vendors supporting the IHE-XDS-I and mammo profiles. I agree that exchanging native DICOM exposes risk of losing some proprietary functionality between systems but, given the advances in PACS technology, the receiving system should be able to execute all of their features and functions if give a fully compliant DICOM object. The mammography exchange will send images into a DICOM based Vendor Neutral Archive that can send the neutral DICOM format or with some basic edge transformation, receive the DICOM in the format of the receiving PACS. This way, all of the tools of the receiving PACS may be used.

Security is and will remain an issue that we should be constantly reviewing and evolving the standards to reflect the most up-to-date technology. Today, sharing a patient's image can automatically trigger a request to the patient to grant or deny permission to share the image, but additionally, the use of object based storage techniques and new data scattering algorithms can provide additional security for data at rest and be more tolerant of internet latency issues.
cheryl whitaker (not verified) @
Great stuff; but I wonder about the claim to be the "first".

Wasn't there a Philadelphia-based "National Digital Mammography Archive"(NDMA) that addressed the issue of cross-site distribution of images for mammography ? See, for example, Mitch Schnall's slides at "http://collab.nlm.nih.gov/webcastsandvideos/ngirsv/universityofpennsylvaniaslides.pdf". I know they ran into trouble with patent litigation with InSite One, but don't know what happened to them after that.

Also, wrt. Alan's comment, the IHE Mammo Display profile is intended to specifically address consistency of features and appearance of mammograms, such that it should NOT be necessary to depend on one vendor's equipment, and viewing of priors from other equipment should be identical (including layout, sizing, perceived contrast, annotation and rendering of CAD marks).

David
David Clunie (not verified) @
I agree that the profile should be able to deal with all images from all vendors, but having direct experience, features inherent to specific vendor workstations seem to be sacrified. For a generalized comparison, there is no problem. For very fine or subtle abnormalities, a problem may exist. In addition, one has to address transmission security and times, whether or not compression has been used and various other specifics that I have had to deal with.
Although these transfers between facilities will probably not be with direct lines, they will probably go through the internet. No matter what your download speed, the problem exists, that others are accessing the same server and time slows (bandwidth is used up), encryption and routers may be a holdup. All of these factors have to be looked at prior to doing total screen/diagnostic cases, unless the old exams are just for an overview. Digital is definitely superior but with the old films, copies were nowhere as good as the originals.
Alan Melton (not verified) @
Click here to close