Paragon Consulting Partners

Healthcare IT Solutions and Results

Enterprise Imaging

By Leave a Comment

Much excitement in healthcare today revolves around the unlocked potential of population-level ‘big data’, which can be leveraged to inform diagnosis and treatment best practices, enable earlier intervention or proactive mitigation of disease, and support the development of new and innovative medical procedures and interventions. 

A luminary teaching and research organization, the National Institutes of Health (NIH) is an excellent example, conducting numerous research initiatives such as disease prevention and treatment, oncology, precision medicine, and neurotechnology, to name a few. Data normalization plays a significant role in supporting these initiatives by creating structured and consistent datasets to that can be leveraged by data mining technologies that enable the processing and analysis of the significant volumes of information required to perform this type of population-level research – a task that would insurmountable if not automated.

Cleaning and Screening

At NIH data analysis begins with the screening and selection of research patients who meet the specific protocol requirements for various ongoing clinical studies. This involves collection and evaluation of patients’ complete medical records including family and medical history, EMR data, imaging records, and much more to identify relevant genetic characteristics, demographic factors, disease profiles, and health statuses, etc. 

The NIH screens thousands of patient candidates, and as such have sophisticated methods to collate, standardize, and analyze the aforementioned data. First, patient identifiers are normalized upon ingestion and a temporary ‘pre-admit’ MRN is assigned to ensure consistency across diverse data sets and facilitate longitudinal analysis of the complete patient record by NIH systems and researchers throughout the screening process. This also ensures candidate data is kept separate from approved, active research datasets until the patient has been officially selected – at which time the MRN is normalized once again to an active ‘admitted’ profile.

The Role of Diagnostic Imaging

Imaging data is a key component to the research performed at the NIH. As such, researchers collect patients’ previous imaging from various outside facilities. Key imaging attributes on imported studies, such as study and series descriptors, are normalized according to NIH policies to enable automated data analysis and simplify the screening process for researchers by ensuring exams hang reliably within their diagnostic viewers for quick and efficient review. 

As well, the NIH often performs additional advanced imaging exams throughout clinical research projects. To ensure this newly-generated data accurately correlates with the patient’s prior imaging history, can be easily inspected by advanced data analysis technologies, and enables efficient review workflows for researchers the NIH also enforces normalization policies at the modality level. 

Because the NIH works with a large number of trainees and fellows the aforementioned normalization of diagnostic imaging exams provides the added benefit of creating a consistent educational environment for teachers and students, supporting structured and replicable teaching workflows.

Leveraging and Developing Artificial Intelligence and Machine Learning

Artificial Intelligence (AI) and machine learning (ML) may be new entrants into mainstream clinical settings, but to the NIH they have been in use for many years already and represent foundational technologies that play a significant role in enabling automatic identification and comparison key data elements across hundreds or thousands of relevant research studies, which would not be possible if done manually. 

However, there are some data elements that cannot yet be fully analyzed by these technologies. For those where bias or variability between reported values or criteria can exist, manual intervention is still required to make appropriate adjustments based on interpretation of the context within which the values were taken. This is the tricky thing about data normalization – it’s often context sensitive and the level of reasoning required to consider the ultimate research question or goal when evaluating data relevance and normalization needs cannot yet be reliably accomplished by machines. 

For this reason, the NIH continues to support the development and refinement of AI and ML algorithms by leveraging their enormous collection of clean and consistent data to build diverse training environments and collaborating with luminary AI and ML technology organizations to support the further development of advanced context-aware data analysis and clinical use cases. 

Anonymization/De-identification

Another critical aspect of the research performed at the NIH is the anonymization and/or de-identification of personally identifiable information (PII).  In order to adhere to patient privacy and human subjects research protection regulations, at times, the NIH wishes to, or is required to de-identify research data by removing or obfuscating identifiable data that could otherwise reveal a patient’s identity.

This might be done to allow the NIH researchers to conduct secondary analyses without additional human subjects’ protections approvals (per 45 C.F.R. 46) or in order to share data with other research collaborators. The NIH accomplishes this through standard de-identification or coding techniques and normalization policies with the goal of scrubbing data to remove identifiers.  

However, the NIH employs a specialized technique that ensures a link persists between the patient’s anonymized/de-identified and identifiable data through a process defined as an ‘Honest Broker’. This ensures researchers have the ability to follow-up with patients for care-related issues and/or can continue to follow outside patients for additional or future research purposes with appropriate safeguards and approvals.

The real value of data normalization

From tracking tumor response to new treatment programs to developing statistical models for a population’s health characteristics and risk profiles, data curation on the scale achieved by the NIH not only requires the collection of vast amounts of scattered information for analysis, but also mechanisms to transform and normalize incoming data to create well-defined and comparable datasets. For the NIH, data normalization enables the extraction of valuable and predictive clinical insights that can be used to improve both clinical outcomes and population-wide health. 

If you enjoyed this post subscribe to our blog to be notified when new articles are published.


By Leave a Comment

Data normalization is an essential activity whenever two or more previously independent facilities are migrated into single, consolidated data repository, as it preserves the integrity of incoming data by avoiding broken links between patient records and problematic collisions between device and study identifiers. 

Jude Mosley, Manager of Information Technology, Imaging at Sentara Healthcare knows this better than anyone. She and her team face this very challenge regularly whenever a new hospital or imaging center is integrated into the 5-hospital network in Norfolk, Virginia. 

Hey, that’s my jacket!

The first challenge is ensuring patient jackets from differing systems are reconciled correctly so that no information was lost or attached erroneously to the wrong patient. Using a master patient index (MPI), Sentara is able to match patients between the source and destination systems and update all incoming studies with the correct patient identifier upon ingestion. This results in a single, consistent set of MRNs across the Sentara network, and eliminates future risk that an old MRN could be mistakenly used. 

The next challenge is ensuring study accession numbers remain unique, and that the migration process doesn’t introduce duplicate accession numbers for different patients. To accomplish this, Sentara employed normalization policies that pre-fixed the accession number and updated the location attribute to a 3-character code representing the source facility for all incoming studies. Not only did this avoid potential collisions between accession numbers, it also added a level of transparency to the migrated data that allowed Sentara to quickly and easily identify its original source. 

The Big Bang theory of modalities

While the migration process provides an excellent opportunity to normalize data on-the-fly, it is often necessary to enforce data normalization policies at the modalities themselves. At Sentara, normalization at the modality level is a well-defined process. With each new modality introduction, update, or site acquisition Sentara ensures that every affected modality adheres to a consistent set of pre-defined policies regarding AE title and station naming conventions. When new modalities are installed, or vendor updates are applied, Sentara requires the vendors to be aware of and sign off on the required data policies to ensure they’re properly applied and thoroughly tested after the installation is complete to maintain adherence.

For larger-scale activities, like a new site acquisition and migration, the PACS team prepares a comprehensive list of all modalities and their new AE titles, station names, and IP addresses, and orchestrates a big-bang update process. While this is no small undertaking, through experience Sentara has refined this process to run like a well-oiled machine. Once complete, Sentara has improved the consistency and supportability of their now-larger infrastructure, and once again ensured that data arrives in a predictable and consistent manner. 

A shining example of normalcy

This case provides excellent examples of how data normalization can address the integration challenges faced by expanding health systems. Not only does it mitigate risk by avoiding data loss and collisions during the migration process, it also measurably improves data quality and reduces support costs in the future by improving the consistency and predictability of systems and data across the entire network.

Our next blog will be taking a more clinical focus, looking at how data normalization can be leveraged to uncover deep insights across population-wide data sources. Sound interesting? Subscribe to our blog to be notified when the next post becomes available!


By Leave a Comment

Normalization has become an essential activity for any data-driven organization. While it does require an investment in time and resources to define and apply the policies, structures, and values you will find that it is well worth the effort. Not only will you see measurable improvements in the quality and efficiency of clinical workflow, stakeholder satisfaction, and your bottom line – you will also unlock the untapped potential of what could prove to be one of your organization’s biggest assets – your data! That being said, let’s take a look at the various methods that can be used to make sure your data falls in line and conforms to your new norms.

What does normal look like anyway?

The first step is identifying and defining the attributes to be normalized. This begins with a close look at your organization’s key challenges and goals. Having trouble with data integrity, fragmentation, and collisions? Take a look at how unique ID’s are assigned, reconciled, and managed at the facility, system, device, and patient levels. Hearing lots of complaints from your clinical team about broken and inefficient workflow? Consider looking at the variations in procedure, exam, and diagnostic naming conventions across your enterprise. Once the troublesome attributes have been defined key stakeholders should be engaged to define what the ‘gold standard’ values should be, and map these to the ‘dirty’ values that are plaguing your system.

Keeping it clean on-the-fly

Once the normalized values have been defined transient systems, such as DICOM and HL7 routers, or destination systems, such as EHR/EMRs, Picture Archiving and Communication Systems (PACS) or Vendor Neutral Archives (VNA), can often be configured to inspect the data as it arrives to dynamically identify and adjust any inconsistencies and ensure what is stored and shared adheres to normalization policies. This is accomplished through a normalization or ‘tag morphing’ rules engines that are able to inspect incoming data on-the-fly, identify deviations from normalization policies using pattern matching algorithms, and apply the appropriate transformations based on the pre-defined set of consistent values. This ensures incoming data is clean, consistent, and reliable – regardless of its original source or format.

As well, it enables rapid integration of outside facilities and systems resulting from mergers or acquisitions as new systems and modalities can be integrated quickly without requiring updates to hanging protocols, routing and retention policies, reporting systems, etc.

Finally, it mitigates the impact of any unforeseen changes that may occur due to vendor updates at the modality, which can sometimes alter attribute values such as series descriptions. This is most common among complex multi-slice acquisition devices, and ultimately results in broken hanging protocols and frustrated radiologists.

Garbage in, garbage out

In some cases, it may be necessary to enforce data normalization policies at the modalities themselves. This is especially important if receiving systems do not provide robust tag morphing capabilities, leaving you without the ability to enforce normalization policies on-the-fly. As well, if you find that your technologists are frequently required to manually enter or adjust data values at the modality, then added variability and potential for error in the resulting data sets are more likely to occur. This may not always be caught by a discerning rules engine. In either case, why not take the opportunity to ensure your modalities are sending clean data from the get-go? As the old adage says: garbage in, garbage out.

When you’re going where the grass is greener

If you’re considering a retiring and replacing ageing systems, data migrations present an excellent opportunity to clean and normalize existing data as it moves between systems, providing immediate benefits like better access and filtering of relevant priors, improved reading efficiency through consistent and reliable hanging protocols, and the ability to incorporate historic data into analytics, AI, and deep learning applications.

As well, it positions you to minimize the effort involved in any future system replacement or migration activity by simplifying the configuration of any system features that rely on specific attribute values or structures to function effectively. For instance, hanging protocols are not typically transferrable between vendors’ systems and therefore need to be re-created whenever a PACS replacement or viewer change occurs. The consistency of normalized data facilitates rapid configuration of protocols within a new system, as the complexity associated with configuring multiple protocols for each distinct layout, or the viewer provided lexicons required to accommodate or ‘map’ the various combinations and permutations of inconsistent protocol, study, or series information is eliminated. The same holds true for other attribute-driven features including but not limited to reading worklists, routing or forwarding rules, information lifecycle management policies, and analytics and deep learning.

That sounds like a lot of work…

More often than not the perceived effort of defining such practices and data structures seems overwhelming to many already busy IT departments and project teams. This often results in a choice to forego data normalization activities in favour of saving some time and effort, or simply due to lack of financial and human resources.

While it may be true that data normalization is no small task, the benefits far outweigh the cost of the initial investment, and many organizations are now realizing the strategic value of data normalization initiatives. The heaviest lift, by far, is the process of gathering and analyzing existing data and distilling it into normalized data structures and values, which is a one-time effort that will yield immediate and recurring dividends by creating an actionable data repository that supports clinical and business continuous improvement initiatives.

By now you might be thinking, “this sounds good in theory, but it’s pretty anecdotal. Where’s the real-life evidence that data normalization is worth the effort?”

Our next posts will include real-world examples of how some luminary healthcare organizations have leveraged data normalization to achieve a variety of measurable benefits. Subscribe to our blog to be notified when the next post becomes available!


By Leave a Comment

A few years ago we signed up for what promised to be an exciting adventure: Extreme Fitness Vacation in the Grand Canyon: Hike from the Rim to the River and Back in One Day! Sounds fun, right? If you think so, or if you’re just curious what this has to do with project management, read on!

This excursion was led by a gym owner, manager, personal trainer, and beefcake in his own right. The trip was seemingly well planned; our routes were outlined and our host promised to take care of all the incidentals – including a relaxing lunch at the bottom of the canyon before beginning the arduous hike back up. What could possibly go wrong? Unfortunately, lots.

To start, it turns out this particular hike (or at least the upwards part of the journey) is actually noted to be one of the 10 most dangerous hikes in North America – not for the faint of heart, and certainly not for the faint-of-fitness. This little tidbit was not known to our excursion leader, who himself was somewhat of a hiking newb, and thus there were no clear fitness or hiking experience guidelines or requirements laid out for those who were interested in signing up. All that was required, as it turned out, was the ability to pay the less-than-modest fee. This resulted in a rag-tag group of pleasant, yet ill-equipped and ill-informed individuals joining the excursion.

The temperature in the Canyon at that time of year (June) often exceeds 110F (or 43C for my Canadian friends), which means a few things: first, you’ll need lots of water and electrolytes (critical to muscle function) because you’re going to sweat. A lot. Second, you won’t feel hungry during the hike, but trust me, your body will be. This means you need to bring lots of light snacks to eat along the way – even when the last thing you feel like doing is eating – to ensure your energy levels are maintained to support your level of physical exertion.

To make a long story short, because of his lack of experience and knowledge of the physical and environmental demands of a Grand Canyon hike, our host made a few critical errors:

  1. He did not bring adequate or appropriate snacks for the number of people in his group (seriously ‘Arnold’s extreme protein muscle fuel bars’ are really the last thing you want to be trying to digest under these conditions)
  2. He assumed there was water along the trail – which would have been great because it meant we wouldn’t have to carry it ourselves! However, water was only available halfway up the return trail, so for 75% of the hike there were no viable re-fill options for empty bottles and thus many among us ran out of water…quickly.
  3. He failed to provide or educate participants on the importance of electrolytes, which are lost with sweat and must be replenished to ensure your leg muscles function well and long enough to haul your butt out of the canyon (as many-a-sign said, down is optional – up is mandatory)
  4. Lunch was promised at the bottom of the canyon, but he didn’t pack anything as he expected he could purchase sandwiches at a river side cafe. The problem? There was no such cafe, and nowhere to purchase sandwiches (unless you want to travel another mile or so off-route to the famed Phantom Ranch, and you’ve booked ahead…)

The result? Heat exhaustion, severe muscle cramping, and even delirium were experienced by some. I’m happy to report we all made it back out without serious injuries – no thanks to our leader, who himself could barely walk due to leg cramps near the end.

So what does hiking the Grand Canyon and Healthcare IT project management have in common? Simple. In this example our fearless (to a fault) leader acted as the project manager. He defined the scope of the excursion, the paths we would take, and the schedule we would adhere to. He was an expert in health and fitness, so the participants trusted his judgement and his plan – and thus followed him on what turned out to be a dangerous and misguided adventure. What he wasn’t, however, was an expert on the specific challenges presented by a Grand Canyon hike and therefore was ill-equipped to identify and navigate the unique terrain and conditions the group encountered.

Proving that even the best laid plans go awry, this (albeit extreme) story provides the perfect parallel for healthcare IT project management. Certified project management professionals (PMPs) are intimately aware of all associated best practices and methodologies required to steer general project initiative and are an important part of any project team. However, unless they also have the subject matter expertise that comes from hands-on experience in clinical and technical healthcare settings they are not equipped with the knowledge or tools needed to craft the most efficient and effective plan or steer the project flawlessly to success.

In addition to knowing how to run a project a PMP who is also a true subject matter expert brings immeasurable added value by additionally understanding the needs of clinical stakeholders, how data flows and systems interact, and knowing what pitfalls may lie ahead – thus allowing them to take proactive steps to help ensure their healthcare IT projects run smoothly, and are delivered on time and within budget.

For more information on our health IT project management methodology download the Six Steps For Effective Health IT Project Management Infographic.

If you enjoyed this post subscribe to our blog to be notified when we post new articles.

 

 

By 1 Comment

‘Enterprise imaging’ is one of the most popular topics within healthcare IT circles today, and it’s no surprise why as within these two simple words lies a lot of potential. Enterprise imaging promises to tear down the silos that traditionally existed between disparate healthcare providers, departments, and facilities. It promises to create a truly unified patient record that is centrally managed and accessible by care providers and patients alike, in real-time. And, it promises to do all this while reducing costs and improving operational efficiency.

Such a state would finally allow patient records to be truly portable, enabling patients to assume ownership of their medical records and empowering them with the flexibility and freedom of choice to navigate the healthcare system with greater ease. As well, care providers would be equipped with critical insights into their patients’ medical conditions and history directly at the point-of-care, allowing them to make timely and informed decisions that ultimately lead to better clinical outcomes and preventative care initiatives.

So why then is Enterprise Imaging still a topic of discussion, and not yet a widely adopted reality?

While the vision is straightforward, the path there is considerably less-so. Every healthcare organization is different, every vendor is different, and it seems that most of the time there are more questions than answers. Should we go with a single vendor that offers a consolidated platform, or pursue a ‘best-of-breed’ strategy that allows us to pick and choose my favourite solutions? Should we update or replace our systems all at once, or take a more progressive approach that leverages our existing infrastructure? Should we host solutions ourselves, or consider a cloud-based and vendor-managed alternative?

Successfully deploying a centrally-managed enterprise platform that intelligently aggregates diverse imaging data from across the care continuum has broad implications and requires a carefully orchestrated approach to solution design that includes a thoughtful analysis of your current situation inline with your unique near and long-term goals.

Our Six Tenets of Enterprise Imaging provides a roadmap that can methodically guide you through the process of developing a complete Enterprise Imaging strategy that is tailored to your unique needs. While it may not provide all the answers up front, it breaks down the key components for success into manageable and actionable steps that consider your workflows, resources, and budget:

  1. Governance: ensures you have the people and processes in place to make informed and timely decisions
  2. Enterprise Platform: addresses the technologies required to build a scalable and performant foundation
  3. Workflow: enables efficient task orchestration and communication between providers and systems
  4. Visualization: delivers unencumbered access to images and tools to care providers across the care continuum
  5. Exchange: allows secure sharing of critical patient information with outside provider or payer networks
  6. Analytics: unlocks the clinical and business insights hidden across your enterprise

To see the big picture, download our infographic. To learn more about each tenet subscribe to our blog, where we will be discussing the key considerations and implications of each in turn.