RamaOnHealthcare April 22, 2022
Today, RamaOnHealthcare talks with Will Perry, Principal, US Cloud Innovation and Engineering Leader, PwC on the bold approach to rare disease.
Introduction: Will Perry, Principal, US Cloud Innovation and Engineering Leader, PwC, discusses how PwC, The Open Source Imaging Consortium (OSIC) and Microsoft are taking a bold approach to rare disease. Through the Open Source Imaging Consortium Data Repository, a first-of-its-kind open source medical imaging and data repository platform, the three parties are highlighting new possibilities to help improve the speed and accuracy of Idiopathic Pulmonary Fibrosis diagnosis and help patients, providers and researchers better manage the disease.
RamaOnHealthcare (ROH): What challenges did you see in the healthcare industry that inspired you to partner with OSIC and Microsoft on this project?
Will Perry (WP): Every rare disease by nature is unique. Whether it is a combination of symptoms, an unusual progression pattern, or a specific genetic mutation, each rare disease typically stands on its own and requires a unique plan for care.
What makes the diagnosis and treatment of any rare disease so difficult is the rarity itself. There aren’t many diagnosed individuals at any given point in time, so often there isn’t enough accessible data for the medical community to learn from. Adding to this, the scarce information available is fragmented across hospitals and clinics around the world, meaning there is a lack of consolidated data on the rare disease that could help doctors and researchers access key patterns to bolster treatment.
…meaning there is a lack of consolidated data on the rare disease that could help doctors and researchers access key patterns to bolster treatment.
Historically, the medical community has focused its attention on rare diseases individually. This means piecing together enough information from individual patients to try to understand the disease at hand. This was the driving force behind PwC’s decision to partner with the Open Source Imaging Consortium (OSIC) and Microsoft to build The Open Source Imaging Consortium Data Repository. We decided to take a new approach to fighting rare diseases by solving the problem of “rarity” itself.
We decided to take a new approach to fighting rare diseases by solving the problem of “rarity” itself.
Instead of focusing on each rare disease individually, the thought behind the data repository is to seamlessly aggregate and share relevant information – like anonymized high-resolution computed tomography (HRCT) scans and appended clinical data – around the world, so researchers could finally start compiling a patient pool large enough for real study.
The first mission for the data repository is to fight Idiopathic Pulmonary Fibrosis (IPF), a rare lung disease with no known cause or cure. With a 3-5 year average life expectancy rate, healthcare professionals historically have grappled with managing the disease.
The repository has already brought together 122 experts and AI-driven diagnostics, transforming the speed and accuracy of IPF diagnosis. In fact, 15,000 scans are projected to be conducted by the end of 2022.
ROH: What are the current clinical challenges of Idiopathic pulmonary fibrosis (IPF)?
WP: Idiopathic Pulmonary Fibrosis (IPF) is a serious chronic disease that affects the tissue surrounding the air sacs, or alveoli, in your lungs. This condition develops when that lung tissue becomes thick and stiff for unknown reasons. Over time, these changes can cause permanent scarring in the lungs, making it difficult to breathe.
A key element to diagnosing IPF lies in identifying a specific pattern on a chest scan via high-resolution computed tomography (HRCT). Identifying patterns on these scans can be challenging. The limited data around the disease often challenges doctors and medical providers are then only left with limited snapshots based on highly variable, siloed information. Creating a more complete picture of what patients face and decreasing the time to diagnosis are more impactful ways to improve outcomes for those with IPF.
ROH: What role does each party have in building The Open Source Imaging Consortium Data Repository?
WP: OSIC first identified the need to create a data repository for IPF, specifically. The team’s blending of radiology, pulmonary, and data science specialists had the right idea, but they lacked the technical experience to bring it to life. That’s when OSIC tapped us at PwC as an implementation lead to design and build this project from the ground up.
From there, we brought in our partners at Microsoft, enlisting the help of its advanced engineering team and Microsoft Azure capabilities that could turn OSIC’s vision into reality.
Using Microsoft Azure, we constructed a data repository to house all of OSIC’s anonymized images and clinical data. The platform looks to incorporate a crowdsourced algorithm that detects IPF progression. This could not have been done without OSIC’s expertise, who led the intake and management process for diagnostic imaging data and performed both manual and automated data quality checks.
ROH: What are some benefits that the repository offers patients, doctors, and providers?
WP: The repository all comes back to one key element: data. We knew from the inception of this project and the problem at hand, that sharing data and knowledge around the disease was critical in making a difference.
That said, a data-powered solution is not successful until people can understand and use it. The platform is designed to dramatically increase the volume and usability of vital information to fuel faster decisions and predict better treatment outcomes.
The platform is designed to dramatically increase the volume and usability of vital information to fuel faster decisions and predict better treatment outcomes.
There are a few vital improvements that this repository provides across the healthcare ecosystem:
- More diverse data: Only limited information on IPF was available and because it is rare, no single organization had enough patients to supply the data needed for real progress. The repository’s large database can help healthcare professionals build algorithms to find answers.
- Worldwide connectivity: The platform consolidates a plethora of global, multi-ethnic data that can help cut down the time to diagnosis.
- Democratizing medicine: Clinicians using the OSIC platform in any location have access to and benefit from the same technology as those affiliated with major research centers. All OSIC-created algorithms are made open source for the benefit of patients globally.
- Continuous improvement: The platform is designed to explain the patterns detected by its AI so providers could then apply their own insights and learn alongside the AI.
All OSIC-created algorithms are made open source for the benefit of patients globally.
ROH: How do you see the repository evolving over the next 5-10 years?
WP: The approach we have taken with OSIC and Microsoft has ramifications well beyond just IPF diagnosis. In theory, such an open-source cloud repository could change the game for any rare disease, because it solves for most rare diseases’ common problem: data scarcity.
There has already been some impressive early feedback stemming from the repository. That has us thinking big about what is next. Its model represents a beacon of hope for those fighting pulmonary fibrosis and other rare lung diseases.
One specific area we see the repository making a difference in is in medical imaging, particularly new uses for old imaging. Solutions in high-resolution computed tomography (HRCTs) aren’t available everywhere, but the OSIC platform could make better use of more widely available X-ray technology by associating patient X-rays with their CT-based diagnosis. In time, the platform could expand to different modalities, including X-rays.
In addition to this, the OSIC platform could help create applications for other diseases. The platform is very flexible and its AI can push the limits of other medical diagnoses beyond IPF. As the database expands and adds more contributors, it’s likely to help spotlight other rare diseases, providing necessary research and attention.
ROH: The data repository is just one example of how modern technologies are advancing the medical industry now. How is AI making a difference in the healthcare space?
WP: The global healthcare AI market is predicted to reach $44.5 billion by 2026, highlighting just how essential the technology has been in advancing the healthcare industry.
The global healthcare AI market is predicted to reach $44.5 billion by 2026, highlighting just how essential the technology has been in advancing the healthcare industry.
AI is helping healthcare organizations apply cognitive technology to unlock vast amounts of health data and power diagnosis. The technology can help clinicians review and store far more medical information – every medical journal, symptom, and case study of treatment and response around the world – exponentially faster than any human.
Similar to the goal of OSIC’s data repository, AI is helping accelerate diagnosis and detection of a multitude of diseases and ailments, as well as consolidating data on those illnesses. It is creating a more widely shared and broader healthcare ecosystem where medical professionals can learn, collaborate, and grow.
About Will Perry
With nearly 30 years of experience in transforming organizations, Will Perry leads by preparing clients to thrive in the future. In his current role at PwC, Will works with clients to create competitive advantages through cloud technology. Will and his colleagues at PwC do this by leading with a business first point of view, based on a multi-competency mindset that challenges and changes our client’s structure, people, process, and technology.