Why you should attend SIIM 2019, June 26-28

One of the major issues facing healthcare imaging and informatics professionals is the lack of

transparency in communication between modalities such as CT, MRI, ultrasound, CR and others, as well as the PACS/RIS, and VNA. When images and related information such as dose information and measurements do not get across, connections are rejected, changes and updates in the information are not propagated in a timely manner or at all, most of PACS administrators are stuck in between vendors who are finger-pointing to each other about the root cause.

Many vendors lock up the access to their log files requiring a (costly) service call to get someone to look at it, which takes time, assuming they have the skills to do that. This has become the main reason why people are attending advanced training classes and why you should consider attending the annual SIIM conference this year.

Imagine that an image is rejected by the PACS or you can’t read images from a CD of a patient who is scheduled for surgery and the surgeon really needs access to the patient’s CT study. The DICOM Validator (DVTK) toolkit will validate the DICOM header and tell you what is incorrect so you can fix it with an editing tool. Imagine that your system randomly loses some of the images in a study. The Wireshark DICOM sniffer will allow you to see exactly what is happening and why images are being rejected. Imagine you have performance issues; the same Wireshark will show you exactly the timestamps of the DICOM communication protocol and application level responses. Imagine there is missing information in a modality worklist, the Mirth HL7 interface engine allows you to map it to a field that could not be recognized by the worklist provider, and also as an extra bonus will store the HL7 orders in a temporary queue, which can be restarted in case there are any hiccups. Imagine the radiology report has some formatting issues, again, Mirth will be able to solve these issues. By the way, all of these tools are free for you to use.

Given the increase in de-constructed PACS, VNA’s that are connected to multiple PACS systems, which require constant synchronization, and the proliferation of zero footprint viewers that can be launched from an EMR, integration is getting more and more challenging requiring complex skills. SIIM leadership has recognized that teaching the advanced skills on how to use these tools will fill a major need and has expanded the program for this year. There will be sessions on using them in a very hands-on manner to provide you with this advanced knowledge.

There are other reasons to attend SIIM as well, i.e. spending time with vendors to kick the tires, learn about the latest in AI, networking with your peers and to share experiences, and last but not least enjoy the great Rocky Mountains. However, all of this is in my opinion minor to acquiring the necessary skills to make sure you can support your PACS in a professional manner. 

So, this is a very good reason to attend this year, I am looking forward to seeing you in Denver at these advanced sessions!

HIMSS19: Are we finally unblocking patient information?

Busier as ever

More than 45,000 visitors to the worlds largest healthcare IT conference held in Orlando, Fla,

browsed through 1200 booths looking for IT solutions for their facilities and listened to the many educational sessions. There is still a dichotomy between what was shown and the real world as the IHE showcase demonstrated 12 use cases where information seamlessly flowed between different vendors, while it is not always so smooth in practice based on stories from the trenches.

Here were my observations from this conference:

Distinguished panel at Keynote

1.       Interoperability, are we there yet? The meeting was dominated by the recent information-blocking rule, which was unwrapped by the US department of Health and Human Services (HHS) literally the day before the convention started. As Seema Verma, the US CMS administrator pointed out in her key note presentation, the government has given out US $36 billion on incentives to implement electronic health records with not much interoperability to show for it, so now it is time for the industry to step up.
Former US CTO Aneesh Chopra added to this saying that the CCD’s (Continuity of Care Documents) that are exchanged right now might not be the best solution to exchange patient information, but we need to look for other means such as open APIs, which can be used to tap into any EMR for information. These open APIs will become a requirement by 2020 according to the HHS. Penalties to health information exchanges and health information networks could be up to $1 million for lacking interoperability. Maybe this will help, however the rule is expected to get pushback from some of the stakeholders. For example, the AHA was quick to point out that it disagreed with certain parts of the requirements: “We cannot support including electronic event notification as a condition of participation for Medicare and Medicaid,” stated AHA Senior Vice President for Public Policy Analysis and Development Ashley Thompson.

2.       Open API, is that sufficient? An open API is merely a “connector” that allows information to be exchanged, however, as was noted in the same keynote speech, if the only thing that can be exchanged is the patient name, sex and race, or if the clinical information is not well encoded and/or not standardized, the API is not of much use.

That is why implementation guides based on use cases specifying the many details of the information to be exchanged are critically important. The good news is that these implementation guides are a key component of the new FHIR standard, which can be electronically interpreted and are defined according to a well-defined template. The DaVinci activity which has already defined 12 of these guides and which are part of the FHIR balloted standard will facilitate the exchange. The focus of these guides is on provider/payer interactions and includes for example medication reconciliation for post discharge, coverage requirements information and document templates and rules. The booth demonstrating these use cases was one of the busiest in the IHE showcase area.

3.       What about social determinants? Health care determinants follow the 20/20/60 rule, i.e. 20 percent of ailments are determined genetically which can increasingly be predicted by looking at your DNA sequence, 20 percent are influenced by a healthcare practitioner such as your doctor, but 60 percent, i.e. the majority, is determined by the patient through his or her own actions and social determinants. For example, if you are genetically at risk for a heart condition, and your doctor has already placed a stent in one or more of your coronary arteries to help blood flow to your heart muscle, but you don’t change your life style, you won’t be able to get any better. Now, let’s say you are homeless and depend on food that is not good for your condition, you could be in trouble. It would be good if your physician knows those social factors, which could also include where you have traveled to recently. However, there are no “codes” available to report this in a standard manner. The majority of the health care determinants (60 percent) are not encoded, therefore, there is much work to be done in this area.

Impressive number of providers
participating in Commonwell

4.       How is information exchanged between providers? The ARRA (American Recovery and Reinvestment Act) from the previous administration had put money aside to establish public Health Information Exchanges (HIE’s). Unfortunately, many of these HIEs took the grant money and folded after that ran out, notably the HIE’s in North Texas and Tennessee and many others, shutdown after failing to find a sustainable business model.
Several vendors took the initiative to establish a platform for information exchange as they figured out that the effort to string connections one-by-one between healthcare providers would be much more expensive than creating their own exchange, which is how the Commonwell non-profit started. As of the conference, they had 12,000 connections to providers, which is probably 10 percent to 15 percent of all providers, which is a good start towards gaining a critical mass. Cerner seems to be the largest EMR vendor in this alliance. Epic was notably absent and has been the main driver of a somewhat competing alliance, called Carequality with different functionality but establishing similar objectives, i.e. exchanging information between EMR’s from different vendors at the providers.
The good news is that there is now a bridge established between these two platforms, which again makes the critical mass even larger. This situation is somewhat unique for the US as other countries have government initiatives for information exchange but for those countries without an initiative, the same model might work. This will hopefully solve the problem that was mentioned by one of the providers who said that it has been relatively easy to exchange information between his EMR (which happened to be Epic) and others as long as it was an EMR from the same vendor, but very hard if not impossible to get anything out of an EMR from another vendor into his EMR. This is a great effort, which together with the anti-blocking rules from CMS, might finally allow healthcare information to be exchanged.

One of the many portals
demonstrated

5.       Are patient portals finally taking off? It is still a challenge to access health care information as there is not really a universal portal that collects all of the information among different providers. You might need to maintain access to the information being present at your primary physician, your specialist(s), your hospital and even your lab work provider. One of the ways to consolidate this information is to have a single provider, such as the VA for veterans, and their portal “myhealtevet,” which has been relatively successful. How this is going to work as the VA  increases its outsourcing to private commercial providers remains to be seen. If you are on Medicare or Medicaid, CMS will provide a standard interface, which is used by several (free) patient portal providers where you can log in to see all of your claims, prescriptions and other relevant information. Again, if you do not happen to be a veteran or are not covered by CMS, but are a patient between the two of these organizations, there is not so much interoperability, however those two groups cover enough patients to start having a critical mass as well.

Standing room only at cloud
providers

6.       Are cloud providers making any progress? One of the speakers, who happened to be working for New York Presbyterian, claimed that they are planning to have 80 percent of their applications and infrastructure in the cloud. There are more advantages to the cloud than potentially reducing cost, such as easier access by patients, and the potential to run AI algorithms for clinical support, analytics and decision support. Machine learning is more effective when there is a lot of data to be learned from, hence the advance of cloud archiving.

The big three cloud providers (Amazon, Google and Microsoft) are more than happy to take your business, the Amazon booth was packed every time I passed it. However, they still have a steep learning curve, even although they claim to have open API’s and a healthcare platform with a lot of features, they have to learn in a few years what healthcare vendors have accumulated in expertise over many decades. The good news is that they have potentially very deep pockets so if they are getting serious about this business, they could become major players.

7.       Is Uberization of healthcare happening? When people talk about Uberization they typically

Get the Uber or Lyft app in your EMR!

refer to the business model that provides easy access by consumers through mobile media, accountability of the providers, and tapping into a completely new source of providers such as private drivers who suddenly become transportation providers, or, in the case of Airbnb, private home owners who become innkeepers.
This is happening in healthcare as well as there is a big increase in tele-health providers who provide phone access to patients who want advice anytime, anywhere. Speaking with a physician who does this from his home, telehealth is as easy as using Uber as a physician can sign up anytime and just log on to take calls from patients for as long as he or she decides. As I listened to one such physician’s experiences, he provided life-changing advice, especially to patients who live in remote areas and otherwise would not be able to seek medical advice because of their remoteness.
In addition to telehealth services, Uber and Lyft were also promoting their transportation business to providers, to reduce potential no-shows by patients who have trouble with transportation. A provider can contract with either one of these to serve their transportation needs.

8.       What about the wearables? Apple introduced access to medical records through its health app

Note the wearable EKG sensor
as well as monitor

at the 2018 conference through a FHIR interface. Their provider list has been growing steadily and is now close to 200, including the VA which by itself accounts for 170 medical centers and more than 1000 clinics. This is a significant number, but not even close to the number of providers in the Commonwell platform, for example, whose members exceed 10,000. Therefore, there is still a lot of progress to be made.
There was an increase in intelligent detectors that can communicate vital signs and other clinical information using blue-tooth to a mobile device, for example, to allow patients to be released earlier from a hospital back home which is safer and more cost effective.

9.       Are we Safe from hackers yet? There has been a major increase in cyber security investment, which has become a necessity as healthcare has become a major target for hackers and ransomware opportunists.

Huge Security pavilion

Healthcare providers are making big investments in personnel and tools to try to protect themselves. The cybersecurity area of the conference was indeed huge as many new companies are providing their services. The average security part of an IT budget is about 6 percent to 8 percent but there are some who spend as much as 12 percent. Imagine an IT staff of a large organization, e.g. if you have 500 IT employees, there could be as many as 50-60 staff dedicated to cyber security.

Based on recent incidents, we still have a lot of work to do in healthcare to protect patient information, especially at the periphery, as medical devices in many cases appear to provide easy access points to a hospital’s back-end. This risk is one of the reasons that the FDA has been requiring a cyber security plan to be filed with every new medical device clearance.

plenty of FAX apps

10.   When are we going to get rid of the fax machines? About six months ago, CMS publicly announced that they want to get rid of all fax machines in healthcare by 2020. This is only one
year away, however in practice it appears that by far the majority of all healthcare communication in healthcare is done through the ubiquitous fax machine. A good manner to transition might be to exchange documents using the open API, and then do Natural Language Processing (NLP) to search for medications, allergies, and other important information that can be processed and potentially imported into the receiving EMR. In the meantime, there were still many small companies that were advertising smart ways to distribute faxes and I predict that this will still happen for several years to come.

In conclusion, this was yet another great conference. The emphasis was on unblocking the information that is locked up in many healthcare information systems and, until now, can only be exchanged if you happen to have an EMR from the same vendor as its source, or if you are lucky enough to have a provider that has access to Commonwell or Carequality, or a provider who uses one of the relatively few public Health Information Exchanges (HIE’s). Hopefully the industry and providers will start to cooperate on making this happen, we’ll see next year when we are going to be back in Orlando FL again.

In the mean time, if you are baffled about some of the terminology you might consider our FHIR, IHE or HL7 V2 training publications and classes.

Cleveland 2019 IHE connectathon: it’s all about FHIR.

The 2019 IHE connectathon drew more than 300 healthcare IT professionals to a snowy and cold

Cleveland to test interoperability between their devices using IHE defined profiles. What was new this year were the recently published profiles facilitating exchange of information using mobile communications based on the new FHIR standard, which is based on standard web protocols.

The FHIR testing emphasized querying for patient information, corresponding documents, and uploading and retrieving them. In addition, we tested the audit trails using FHIR, which is important from a security and privacy perspective.

There are still relatively few FHIR based implementations in the field due to the immaturity of the standard (most of it is still a draft), the lack of critical mass (implementing FHIR for just one application such as scheduling does not make sense), and its steep learning curve as it is sufficiently different from what healthcare IT is accustomed to. Therefore, the more that can be verified and tested in a neutral environment such as the connectathon the better it is. Speaking with the participants, they uncovered quite a few issues in their early FHIR implementations, which again is good as it is better to solve these issues beforehand than during an actual deployment.

The attendance at this year’s event seemed to be lower than previous years, which could be due to the fact that there are several local connectathons happening during the year on other continents, which draw from the same crowd, and that implementations are starting to mature (except for FHIR of course) and therefore, there is less need for debugging.

There could also be somewhat of a “standards overkill” in place if one considers the fact that for radiology alone, which is one of the 11 domains, there are 22 defined profiles and another 27 published as draft. It is hard to keep up with all of these and for vendors to deploy all of these new requirements.

As FHIR matures, which will take several years as its typical release cycle is 12-18 months, there will be a need to test these new releases between the vendors providing the medical devices and software products. There were a few new vendors present but most of them consist of the same “crowd,” which is somewhat disappointing because I believe that the real FHIR implementation breakthrough will come from outsiders such as Apple, Amazon, Microsoft and/or Google, all of which were notably absent.

In conclusion, another very successful event, giving a boost to interoperability, something we very badly need as patients, especially in the US where we are still struggling to get access to our medical records, including images, and where healthcare institutions continue to have difficulty exchanging information among themselves and other providers. In many departments and medical offices, the fax machine still is an important tool, hopefully not for long.

RSNA2018: What’s in and what’s out.

Let it snow...

The annual radiology tradeshow at McCormick Place in Chicago started with a little hiccup as the Chicago airports closed down on Sunday due to a snowstorm, and slowed the flow of attendees flying in on the second day to only a trickle. Note that the Sunday after Thanksgiving is the busiest travel day of the year so it could not have come at a more inconvenient time. I myself was caught in this travel chaos as I spent all of Monday in the Dallas airport while my plane was trying to get into the arrival queue for O’Hare.

The overall atmosphere at the show was positive, attendance seemed to be similar to last year and most vendors I talked with were optimistic. About a third of the attendees come to the meeting just for the continuing education offerings, but another third come to visit vendors and “kick the tires” and see what’s new. My objective is also to see what the new developments are and to do some networking to get an idea of what is going on in the industry. 

Here are my observations:

1.       Artificial intelligence dominated the floor - Over the past few years, AI has created some

Dedicated area just
for AI showed
80 companies

anxiety as predictions that AI would replace radiologists in the near future. It seems that the anxiety has been relieved to a certain degree, but it has been replaced with a great deal of confusion of what AI really is, and with uncertainty of what the day-to-day impact could be.

A detailed description of the different levels of AI and the main application areas are the topic of an upcoming blog post, but it was clear that the technology is still immature. Despite the fact that there were 100+ dedicated AI software providers, in addition to many companies promoting some kind of AI in their devices or PACS, only a handful of them had FDA clearance. I also believe that the true impact of AI could be in developing countries that have a scarcity or even total lack of trained physicians. It is one thing to improve the detection by a physician of let’s say cancer by a few percent, but if AI could be used in a region that has no radiologists, then an AI application being used that can detect certain abnormalities  would be a 100% improvement.

There could be some workflow improvements possible using AI in the short term, however, one should also realize that the window between conception and actual implementation could be 3-5 years. Users are not too anxious to upgrade their software unless there is a very good reason. So, in short, the AI hype is definitely overrated and I believe that we’ll almost certainly have autonomous self-driving cars before we have self-diagnosing AI software.

a significant dose reduction
 for lung cancer screening

2.       Low dose CT scanning is becoming a reality - One of the near-term applications of AI allows the use of a fraction of a “normal” CT scan. Instead of a typical 40 mAs technique, acceptable images are created using only 5 mAs. This could have a major impact on cancer screening. The product shown did not have FDA clearance (yet) but there is every reason to expect that this can be available one year from now. The algorithm was created using machine learning from a dataset of a million images to identify body parts in lung CTs, and subsequently reduce the noise in those images, which allows for a significant dose reduction, claimed to be 1/20

Extremity
Cone Beam CT

3.       Cone beam CT scanners are becoming mainstream - Cone beam CT scanners were initially
used primarily for dental applications where the resulting precision and high resolution images, especially in 3-D, are ideal for creating implants. However, for ENT applications, such as visualizing cochlear implants and inner ear imaging, its high resolution and relatively low cost makes them ideal. It is also very useful for imaging extremities, again, its high resolution can show hairline fractures well and is superior to standard x-ray. I counted at least 5 vendors offering these types of products; they are being placed in specialty clinics (e.g. ENT) as well as large hospitals.

4.       Point-Of-Care (POC) ultrasound is booming - POC ultrasound is getting inexpensive (between US $2k-15k), which is affordable enough to put one in every ambulance, and in the hands of every emergency room physician, and even for physicians doing “rounds” and visiting bedsides. There are different approaches for the hardware, each with its own advantages and disadvantages:

a.       Using a standard tablet or phone, there is an “app” needed for the user interface, image display, and upload to the cloud and/or PACS. All of the intelligence is inside the probe. However, one of the complaints I heard is that the probe tends to be somewhat heavy and can get very warm.

b.       Using a dedicated tablet modified for this use, it can take some of the load off the probe
for the processing. If the probe is powered through the tablet, it saves on weight as well.

Butterfly POC US, US$2k

Other things to look for is whether a monthly fee is included as several vendors use a subscription model, if it has a cloud based architecture (i.e. no stand-alone operation), and what applications can it be used for. Most of the low-end devices are intended for general use, and have only one or two probes. If you need OB/GYN measurements, you might need to look for a high end (close to US $10k-15k price range).

Also, uploading images into a PACS is nontrivial as one needs to make sure it ends up in the correct patient record of the PACS, VNA, EMR, etc. This is actually the number one problem as each facility seems to deal with these so-called “encounter-based” procedures in a different manner. There are guidelines defined by IHE, but in my opinion with a very narrow scope.

5.       3-D Printing is becoming mainstream - A complete section at the show was dedicated to 3-D  with regard to X3D/VRML models in ongoing. So, before you make major investments, I would make sure you are not locked into a proprietary format and interface.

Many companies showing off
printed body parts

printing. Several vendors showed printers and amazing models based on CT images. The application is not only for surgery planning (nothing better than having a real-size model in your hands prior to surgery) but also for patient education to share a treatment plan. I would caution however that the DICOM standard (as of 2018) includes a definition on how to exchange so-called “STL” models, but the work

There is not (yet) a large volume of these printed models. I talked with a representative of major medical center, who said they do about 5-10 a day, and another institution, i.e. a children’s hospital does about 3 per week. It seems to me that creating orthopedic replacements might become a major application, but then we ae not talking about models you can make with a simple printer that creates objects from nice colorful plastic, but rather one that can compete with the current prosthetics based on titanium and other materials.

6.       Introduction of new modalities - Every year there are several new modalities introduced, which are very promising and could have a major impact on how diagnosis is done in a few years for particular body parts and/or diseases. Examples are a new way to detect stroke by using

Dedicated Breast CT

electromagnetic imaging for the brain. The images look very different from a CT scan, for example, but it gives a healthcare worker the information they need to make treatment decisions. Another new device is a dedicated breast CT device providing very high resolution, 3-D display and is more comfortable for a woman than a regular mammogram. Note that these devices don’t have FDA clearance (yet), but as common for these new technologies, they are deployed in Europe and as soon as the FDA feels comfortable, they be ready for sale in the US as well. On issue with these devices is that there is no real “predicate” device so they need clinical trials to show their benefits.

Equally important to what’s new is also observing what’s “old,” because the technology has become mature, or it has made it beyond the “early-adopter” stage. This is what I found:

1.       PACS/VNA/Enterprise imaging - Over the past few years, PACS systems have become mature and not much talked about. Most investments by institutions have been with new EMR’s so there has not much left over to upgrade the PACS system. The result is that many hospitals run several years behind in upgrading and/or replacing their PACS, which hurts the most when needing to facilitate new modalities such as the breast tomo (3-D) systems. One is forced to stick with proprietary solutions to make these work and/or using the modality vendor’s workstations to view these.

VNA implementations have also been spotty. Some work rather well, but some have major scaling and synchronization issues between the PACS and VNA. Enterprise imaging was touted the past 2 years as well, but as a result of a lack of orders (see discussion above about POC ultrasound) creating work-arounds, has not really taken off as expected. New features are needed such as radiation dose management, peer reviews, critical results reporting, and sophisticated routing and prefetching, which are solved by using third party “middleware” to resolve these issues.

2.       Blockchain - Using blockchain technology in healthcare has a limited application. The reason is that the bulk of the healthcare information does not lend itself to be stored in a public “ledger.” It is nice that the information cannot be altered, but unless it is completely anonymized (which is still an issue as there can be “hidden information in private data elements, embedded in the pixels, etc.), and made available for research purposes for example, there are not that many uses for this technology. As of now, some limited applications such as physician registries seem to be the only ones that are feasible in the short term.

3.       Cloud solutions - Google, Amazon and Microsoft are the big players in this market, but there are still very few “takers” for this technology. One of the reasons is the continuing press on major hacking events into corporations (500 million records from Marriott hotels is the most recent as of this writing) and reports of ransomware events of hospitals. Even though one could argue that the data is probably safer in the hands of one of the top cloud players than on some server in a local hospital, there is definitely a fear factor.

As an illustration, one of the participants told me that their hospitals cut off all of the external communications, so there is no Internet at all on any hospital PC. I have seen many physicians Googling on their personal devices such as tablet or phone instead, to search for information about certain diseases or cases. Despite the push from Google et al we probably need some real success stories before this becomes mainstream. Note that what I call “private cloud” solutions, which are provided by dedicated medical software vendors, are doing better, especially for replacement of CD image distribution and for allowing patients to access their images.

Overall, there was quite a bit to see and listen to at this year’s RSNA. Because of the weather cutting into my visit, I was barely able to cover everything I wanted to during the week. It was interesting to see how mature image processing techniques suddenly appeared as “major new AI” solutions, how there are still so many in their infancy, which makes me to believe that the immediate impact will be relatively little. I was more excited by new modalities and inexpensive ultrasounds, which will have a major impact. 

I am hoping that next year some vendors will spend more effort going back to some of the basics, providing robust integration and workflow support for the day-to-day operations. We’ll see what will be new next year!

SIIM18 meeting: Is AI reality yet?

The Artificial Intelligence (AI) hype from the RSNA meeting in Chicago definitely spilled over to the SIIM meeting held at the National Harbor in DC, May 31-June 2, 2018. There were several new upstart companies that were showing various new algorithms being applied to medical images and there were quite a few presentations about this subject.

Here are my top observations on this subject:

·        AI is nothing new - As Dr. Eliot Siegel from the VA in Baltimore said at one of the sessions. “I use AI all day, when I use my worklist, when I do image processing, or when I apply certain calculations; I have been doing that for several years before the term AI was coined.”

·        The scope of AI is continuously changing - as pointed out by the anonymous Wikipedia contributors on the definition of AI, what was considered AI technology several years ago, e.g. optical character recognition, is now considered routine; in other words, “AI is anything that hasn’t been done yet.”

·        Even the FDA realized that CAD (a form of AI) is becoming a mainstream, mature technology. The FDA has proposed reclassifying what it calls radiological medical image analyzers from class III to class II devices. The list includes CAD products for mammography for breast cancer, ultrasound for breast lesions, radiographic imaging for lung nodules, and radiograph dental caries detection devices.

·        AI can determine which studies are critical - With a certain level of confidence, AI algorithms can distinguish between studies that very likely have no finding and those that require immediate attention and sort them accordingly. Note that this requires the AI software to be tightly integrated with the workstation worklist that drives the studies to be reviewed for the radiologist, which could be challenging.

The "AI" domain name has become
popular among these early
implementers

·        There are many different AI algorithms, and none of them are all inclusive (yet) - If you would take all of the different AI implementations, one might end up with maybe ten or more different software plug-ins for your PACS, each one looking for a different type of image and disease. Even for one body part an AI application does not cover each finding, for example, looking at a vendor’s chest analysis, it listed 7 most common findings, but it did not include the detection of bone fractures.

·        What about incidental findings? - The keynote speech at the SIIM was by e-patient Dave who made a very compelling case for participative medicine, i.e. partnering with your physician, being possible by sharing information and using web resources. His story started with an incidental finding of a tumor in his lung which happened to show up in a shoulder X-ray. If this image was being interpreted by AI that was only looking for fractures, his cancer would have been missed, and he would not have been here today.

·        There is no CPT code for AI - This leads to the question of how to pay for AI. Especially for in-patients, for whom additional procedures such as processing by an AI algorithm are an additional cost. Any extra investment and/or work needs to have a positive return on investment. This would be different of course if AI can improve efficiency, accuracy, or has any other measurable impact on the ROI.

Example of Presentation State
display on image

·        Consistent presentation of AI results is a challenge - AI results are typically presented in the form of an overlay on the image and/or in combinations of key images indicating in which slices of a CT, MR or ultrasound study certain findings are shown. These overlays are either created in the form of a DICOM Presentation State (preferably color) or, if there is no support for that, as additional screen saves with the annotations “burned” into the pixel data, both appearing as separate series in the study and stored on the PACS. A couple of AI vendors noted the poor support by PACS vendors of the color presentation states as several of those apparently changed the display color upon presentation on the PACS workstation.

·        Few vendors display the accuracy - It is critical for a physician to see the accuracy or confidence level of the AI finding. However, as noted in one of the use groups, accuracy is more than just sensitivity and specificity, and there is no standard for that, i.e. how would one compare a certain number between two different vendors?

The definition of AI is being debated, some prefer to call it Augmented or Assisted Intelligence. Some argue that it is nothing new, and indeed, in practice the definition seems to be shifting towards “anything new.” Implementations are still piecemeal, covering relatively small niche applications.

As with self-driving cars, or even auto-pilots in a plane, we are far from relying on machines to perform diagnosis with a measurable and reliable accuracy. In the meantime, for routine tasks AI could provide some (limited) support. An example is for TB or mammography screening, where an AI algorithm could determine that with 99.999 % accuracy there is no finding. The question is what to do with the 0.001 % and with incidental findings, which could become more of an ethical than technical issue.

HIMSS 2018: Wake-up call for the sleeping giants.

As I browsed through the vendor exhibits among the more than 45,000 healthcare IT professionals gathered in Las Vegas last week for HIMSS 2018, I noticed that the big IT giants Amazon, Google, and Microsoft (Apple was noticeably absent from the exhibit floor), as well as other businesses who are in the CRM space (Salesforce) are finally taking notice of the opportunities in healthcare. It was also not a coincidence that Eric Smidt, past chairman of Alphabet, Google’s parent company, was the keynote speaker for the conference. I believe that this is very promising as healthcare in many ways is very much behind other industries and can learn from their experiences.

As anecdotal evidence of the need for better technology in healthcare, I listened to a presentation from a vascular surgeon who explained how he annotates relevant images on a PACS viewing station, then takes a picture with his iPhone of the screen and shares it using Chat with his residents and surgical team to prep for surgery. The reason for him to have to use his phone, is that we don’t yet have the “connectors” that tie these phones, tablets, and other smart devices with our big, semi-closed healthcare imaging and IT systems. The good news is that Apple just announced an interface allowing information exchange, which can be used, among other things, for patients to access their medical information from a hospital EMR. Also Google cloud announced an open API.

Here are my top observations from HIMSS2018:

Demonstration of new Apple
App accessing health records

·       Patients are taking control of their medical information: Apple announced a FHIR based interface on the iPhone that provides access to personal health records. The interface is built into the recent Apple phone as part of its health app. Information such as recorded allergies, medications, lab results, etc. is copied to the person’s phone. Note that this is different from solutions where this information is stored in the cloud (e.g. Google, etc.).

Regardless, it allows patients to access and keep their own information. It provides a mechanism for patients to share the information, as the hospitals are struggling to meet that demand (only one out of three hospitals can share information according to a recent AHA study, despite the fact that more than 90% of them use electronic health records).

In reverse, it is not that hard to upload this information back into an EMR of a physician or a specialist, together with information collected from blue-tooth enabled blood pressure, pacemaker, insulin pump, and other intelligent healthcare devices as well as wearables. At the IHE interoperability showcase demonstration areas, there were several demonstrations of how this upload can be achieved using standard interface protocols, often using FHIR.

·       FHIR is gaining more traction: The new HL7 protocol allowing easy access, especially by mobile devices, to so-called resources such as lab results, reports, and also patient information is getting more traction. However, there is a still a big disparity between what is shown as “works in progress” such as the demonstrations at the IHE interoperability showcase, and what is actually deployed. Almost every use-case that was demonstrated at the showcase had one or more FHIR elements, such as used for patient information access, uploading images or labs, accessing registries, etc. However, when I asked vendors on the exhibit floor where they deployed the FHIR interface, many of them told me that yes, they have their FHIR interface available but are still waiting for the first customer to actually use it.

There are a couple of exceptions, for example, at the Mayo Clinic they are using FHIR to access diagnostic reports, utilizing the EPIC FHIR interface, but there are still very few. One of the major obstacles with FHIR implementations is that it took them a long time (5 years to-date) to get to a standard that has at least some normative parts in it, which will be release 4 to be balloted soon, which means that any implementation you do right now is subject to changes as upgrades are non-backwards compatible. As an example, the Apple FHIR interface is based on release 2. So, I am officially upgrading my FHIR implementation status from “very limited” to “spotty,” but I believe that there is definitely a lot of potential.

Demonstration of VA to DOD gateway
based on FHIR technology

·       The VA is making major strides in healthcare interoperability: I feel compelled to call out the US department of Veterans Affairs as there is a push to shift some of their care to the private sector, while the fact of the matter is that research shows that the VA scores higher than the industry in many of the quality scores, despite the fact that yes, there is still a lot of disparity between the different VA facilities. The high quality of care is not in the least caused by the early implementation of electronic medical records and the ability to be paperless. But, their current medical record system is becoming out-of-date, hence the intention to replace it with a new EMR at the cost of about $10 billion over the course of the next 10 years. Nevertheless in many ways, their current system still outshines what can be achieved today by commercial vendors.

As a case in point, there is a connection between the VA EMR and the one from the DOD that allows for a smooth transition of veteran data between these two entities, which is based on FHIR. What is significant, is that of the many FHIR resources that FHIR has defined (more than 100 up to now, planning to be at about 150), the VA is able to exchange all of the information needed with only very few FHIR resources, notably Patient, Imaging Study, Questinonnaire, Observation, Clinical Impression, Diagnostic Report, Encounter, Condition, Composition, Allergy and Medications. This means that implementing a relatively limited subset can still be very effective. Hopefully their replacement EMR (Cerner?) will have the same kind of interoperability, which seems to be a point of contention right now in the contract negotiations for replacement.

·       The big EMR companies are doomed (or are they?): This millennium has shown a major shift in healthcare IT as the past ten years the number of hospitals in the US having an electronic record has gone from 10% to more than 90%.

However, these monolithic, semi-closed systems which accumulate all the patient information in big databases that are hard to access with limited tools for dashboarding and quality metrics, and who often charge a hefty fee to provide yet another interface to get information in or out, might be on their way out unless they change their architecture and focus. For what it’s worth, even the White House is taking notice as Jared Kushner mentioned during the meeting that “Trump has a new plan for interoperability.”

Let’s look at an analogy on how other industries solve the information access problem, for example, a website for a hotel. If you would like to find directions to the hotel, you click on a link to Google Maps, if you want to know what the local sightseeing tours are, you click on “tripit”, for reviews you click on “Tripadvisor”, and so on.

Now let’s go back to our ideal EMR user screen, wouldn’t it be nice if you can get the patient information from a “source of truth,” which is a web-accessible source for patient information, the latest lab results from the lab, either internal and/or external, the past 6 months progress on a weight loss program from the patient’s Fitbit located in the cloud, diagnostic reports from the radiology reporting system, and so on. And by the way, arranging transportation for the patient is just another click on the Uber or Lyft App (note the announcement from Allscripts to embed a Lyft interface to their EMR).

The EMR would be a mash-up of multiple resources accessible through standard protocols (FHIR), in some cases guaranteed immutable, using blockchain technology, and the only functionality left would be a temporary cache and workflow engine that guides health care practitioners through their job in a very easy to use manner.

Currently user friendliness, especially, still leaves a lot to be desired, as a recent study showed that during an average patient visit, providers spent 18.6 minutes entering or reviewing EHR data on digital devices, and only 16.5 minutes of face-to-face time with patients. We’ll see what happens over the next 5 years and who will win and who will lose but it appears that FHIR might facilitate a disruptive development.

Standing room seats only for blockchain
presentations

·       Blockchain has some (limited) applications in healthcare. I purposely did not mention blockchain in the title of this write up so as not to overload my ISP as I found it to be the most hyped (according to the dictionary: “extravagant or intensive publicity or promotion”) subject of the conference. Presentations on this subject went beyond standing room only.

What is blockchain? It is an immutable, decentralized public ledger that could be used to securely share transactions without a central authority. Knowing that most of the patient’s health information is not intended to be public, and that some of the files (think a 1.5GB digital pathology slide) are just too big to simply move around and copy multiple times, it makes the application for blockchain very limited in scope. The immutable aspect is also hard to accomplish, even for objects or entities that you might think are immutable such as a patient/person.

Imagine that you would store the patient information in a blockchain (e.g. a url and “fingerprint” or “signature” of the data), can you really guarantee that there would be no changes? Some of the content might need to be updated such as a “disease status” in case someone dies, a different name in case a woman who marries, and it is not uncommon anymore for a patient to change sex.

Apart from the “content,” the structure might change as well, due to database changes such as allowing storage of multiple middle names, aliases, etc. Some of these solutions such as providing a unique, immutable person identification, will be resolved by other industries anyway as financial institutions have a lot of interest in making sure that they provide credit to “real persons” and identify if a financial transaction is requested by the actual person instead of a hacker or intruder.

There are however a few blockchain candidates for healthcare, one example was shown at the recent RSNA show dealing with certification and accreditation of physicians, which should be public and from a reliable source. Another example is dealing with consents, so that a healthcare provider can trust the fact that patient information can be shared with for example a parent or caretaker, and what part of the record can be shared and what not (e.g. limit access to mental illness records or the fact that a 16 year old daughter uses contraceptives). So, in conclusion, yes there are some limited applications for blockchain technology, many of them we can “borrow” from other industries, and some of them we can implement for medical purposes, but in practice it will be few.

Salesforce: Patients are
customers?!

·       Healthcare is learning from CRM companies: According to one of the major CRM companies, Salesforce, Customer Relationship Management (CRM) is a technology for managing all your company’s relationships and interactions with customers and potential customers. Replace the word “customers” with “patients” and you have a perfect system that allows a healthcare institution to manage their patients in a better manner. That is why not only Salesforce but other companies (I saw a demo at Microsoft) are using the CRM core to provide patient management solutions.

·       Artificial Intelligence is making small progress: It would not be right not to mention AI in this report as it is in the top ten tweets about the conference. However, machine learning and Artificial Intelligence is still not as easy as one might think. Some researchers indicate that the IQ of intelligent machines to be equivalent of a 4 your old right now. But, as of now, machines are unbeatable for chess and jeopardy, so there are definitely some applications that can benefit from AI. Examples are predicting ER re-admission rates of certain patients and taking action accordingly, assisting a physician to make a better diagnosis, or, even better, ruling out any findings with an almost 100% accuracy, which would assist in routine screenings. In addition to the technology having to become more mature, there is also an issue with data access as I talked with one user who is in charge of entering manually textual data from old records in structured format, and the fact that much of the accessible data is not very structured. There is a lot of emphasis on AI, so much that some companies are re-branding their whole healthcare business around it (think IBM: Watson Health), which also seems an overkill to me. But AI will silently enter into many applications where it can impact workflow, enhance diagnosis and clinical outcomes.

Yes, I want theVespa

·       HIMSS is still an IT tradeshow: Imagine walking around the RSNA (radiology conference) and being asked if you want to enter in a $200 drawing, participate in a magician performance or, enter a drawing for a motorcycle. It would be unthinkable, but it is still common at the HIMSS. This indicates that it gears towards a different audience than clinicians. In contrast with the last time, however, I did not see any showgirls on the floor this year for photo-ops, so the only decision I had to make was if I would enter the motorcycle or scooter drawing. Having driven a Vespa myself when I was young, it was not a hard choice for me.

In conclusion, this was another great event, with some hype as usual, but I found especially the promise of “outsiders” getting involved in the business of healthcare to be very encouraging. A “fresh look” from these companies using some of the practices that make our life easier when we are not sick, could definitely make our life easier and improve patient care when we are sick. There is no reason that financial transactions can freely move between banks so that I can go to an ATM any place in the world and access my account, while my physician has trouble getting timely lab results, medications, allergies and other pertinent information. I can’t wait for the sleeping giants to not only wake up but get actively involved and make an impact.

Herman Oosterwijk is a healthcare imaging and IT trainer/consultant. In case you like to learn more about new standards, in particular FHIR, check out the upcoming web training and in-depth face-to-face training.