Greg Freiherr, Industry Consultant

Greg Freiherr has reported on developments in radiology since 1983. He runs the consulting service, The Freiherr Group.

Sponsored Content | Blog | Greg Freiherr, Industry Consultant | Archive Cloud Storage | November 08, 2019

BLOG: Uncovering Patterns in Big Data

Embedded in the many data sets that comprise it, Big Data may provide an understanding of how health care can be improved

Graphic courtesy of Pixabay

Embedded in the many data sets that comprise it, Big Data may provide an understanding of how health care can be improved. But because its volumes of information can overwhelm traditional means of analysis, valuable patterns may emerge only through the use of deep learning (DL) algorithms. This form of artificial intelligence can transform Big Data into actionable information, giving providers the insights through which they might reduce health care costs and improve care.

With its DL algorithms, the Imalogix platform exemplifies how Big Data can be harnessed to impact health care.

“You can see how you — yourself — are doing or you can see how you are doing versus your peers,” said Thomas Griglock, Ph.D., chief diagnostic imaging physicist at Oregon Health and Science University (OHSU). “You can find out if your doses are high for some types of CT protocols compared to the national average; how the uptime on your scanners compares; how your machines are being utilized compared to others.”

Improved patient positioning during CT, for example, can be achieved through DL analyses. DL algorithms may be leveraged at a single site or across multiple sites within an enterprise to improve image quality and reduce patient radiation dose. This can impact diagnostic accuracy and make monitoring more effective, just as it can improve patient safety.

Physicians may use DL analyses, for example, to quantify the amount of radiation to which patients are exposed during a certain procedure. This may help drive behavioral change that may lead to the lowest possible doses that can be applied while still producing acceptable exam results.

If achieved across an enterprise, DL analyses of Big Data may increase consistency and efficiency. This may boost patient satisfaction by making scheduling more reliable. Increased efficiency may also add to the facility’s financial bottom line.

By leveraging advanced analytic solutions that leverage Big Data and DL algorithms, providers can quickly apply what is learned from those analyses to get continuously better – clinically, operationally and financially. The Imalogix system, Griglock said, “provides real world, actionable feedback. (It is) as close as you can get to plug-and-play.”

Unleashing The Potential Of Big Data

Achieving this potential requires leveraging the cloud. Through the cloud, data sets are collected from multiple sites and transformed into Big Data. Through repeated dives into this data, DL algorithms learn to identify complex patterns. These analyses are then turned into actionable information about how performance can be more effective and efficient; how equipment can be maximized to ensure high throughput without comprising quality; and how high-quality images can be generated while patients are exposed to less radiation.

Imalogix DL algorithms can provide details about equipment utilization and staffing at a single site or enterprise, breaking down information into intuitive dashboards and graphs. (These may be broken down further by hour, site and number of exams.) Or analyses may compare CT techniques performed at one hospital to those performed at ones across the U.S.

At OHSU, Griglock and colleagues use Imalogix analyses to compare scanner operations in the enterprise to each other, as well as to ones at other institutions across the U.S. In these ways, the university uses analyses of Big Data – these comparative analyses -- to identify areas in which operations may be improved. Predictive analyses, according to John Heil, a founder and CEO of Imalogix, may be used to forecast how staff changes can lead to improved performance.

OHSU also uses DL analyses to meet everyday requirements, for example, to comply with mandates. Griglock and colleagues use the cloud-based system to keep tabs on patient exposure to radiation during CT exams to meet requirements set by the Joint Commission.

“The Joint Commission mandated a few years ago that everybody needed to keep track of dose,” Griglock said. “This (the Imalogix system) is probably the best way that I have seen to achieve compliance with those regulations.”

How Imalogix Leverages Big Data

Importantly, Big Data is not only processed in the cloud to provide actionable analyses, it is how DL algorithms learn to uncover actionable patterns. Just as DL algorithms are trained on this Big Data, they continue to learn during their analyses as new information is added.  This allows the algorithms to improve continuously, bringing new discoveries to users.

The goal is not just to use Big Data efficiently. It is to make Big Data usable — “to deliver insights that support new opportunities in cost savings and better patient care,” Heil said.

Analyses using Imalogix DL algorithms may identify opportunities that a site may have to improve patient safety and image quality, for example, by identifying improvements in patient positioning. An enterprise will receive an overall score as to how well its CT technologists center patients in the scanner. That score is compared to those of peers. Users of the Imalogix system receive information on how to improve their scores, information that can benefit the overall team as well as specific technologists.

Identifying areas of opportunity is the first step toward improved performance. The next is to use DL algorithms to analyze Big Data so as to forecast how staff actions might improve the site’s overall performance compared to that of its peers.

“Based on last month’s performance, we predict how well you will do if you followed through with any of the recommendations Imalogix uncovers,” Heil said. “That really gets to the answers of how to change a business and what the performance impact is going to be.”

This process, he said, is never ending: “We engage with our customers daily, watching the data, learning how their businesses are going, figuring out how to help by delivering intelligent and actionable insights.”

Underlying it all is Big Data.


Editor’s Note: This is the third in a four-part series about the value of the cloud in today’s health care. The first blog, How Healthcare Can Benefit From the Cloud, can be read here. The second blog, What AI Does in the Cloud, can be found here.

Related content:

BLOG: How Healthcare Can Benefit From the Cloud 

BLOG: What AI Does in the Cloud

VIDEO: How Imalogix Uses AI to Boost Performance

How Two Providers Use The Cloud To Prepare For Disaster 

Related Content

The U.S. Food and Drug Administration (#FDA) authorized marketing of #Medtronic's #GIGenius, the first device that uses #artificialintelligence (#AI) based on #machinelearning to assist #clinicians in detecting #lesions (such as #polyps or suspected tumors) in the #colon in real time during a c#olonoscopy.

The GI Genius intelligent endoscopy module works in real-time, automatically identifying and marking (with a green box) abnormalities consistent with colorectal polyps, including small flat polyps.

News | Artificial Intelligence | April 12, 2021
Varian announced it is collaborating with Google Cloud to build an advanced artificial intelligence (AI) based diagnostic platform to aid in the fight against cancer. Varian and Google Cloud AI embarked on a deployment journey, using Neural Architecture Search (NAS) technology via Google Cloud AI Platform, to create AI models for organ segmentation — a crucial and labor-intensive step in radiation oncology that can be a bottleneck in the cancer treatment clinical workflow.
News | Artificial Intelligence | April 08, 2021
April 8, 2021 — Varian announced it is collaborating with Google...
Ultrasound is an invaluable diagnostic tool for the early detection of breast cancer, but the classification of lesions is sometimes challenging and time consuming. Could artificial intelligence hold the answer to solving these problems? Graphic courtesy of Chinese Medical Journal

Ultrasound is an invaluable diagnostic tool for the early detection of breast cancer, but the classification of lesions is sometimes challenging and time consuming. Could artificial intelligence hold the answer to solving these problems? Graphic courtesy of Chinese Medical Journal

News | Artificial Intelligence | April 06, 2021
April 6, 2021 — In 2020, the International Agency for Research on...
Nano-X Imaging Ltd (Nanox), an innovative medical imaging technology company, announced that its single-source Nanox.ARC digital x-ray technology has received 510(k) clearance from the US Food and Drug Administration.
News | X-Ray | April 02, 2021
April 2, 2021 — Nano-X Imaging Ltd.
Elsevier’s STATdx, a leading radiology diagnostic decision support solution, now includes select Merit-Based Incentive Payment System (MIPS) Measures validated by MDinteractive
News | Clinical Decision Support | April 02, 2021
April 2, 2021 — Elsevier, a global leader in research publishing and information analytics, announced a partnership w