Supporting Digital Innovation in Radiology, Imaging

March 22, 2021 - In 2016, a team of researchers surmised that 90 percent of all healthcare big data came from imaging. Nearly five years have passed, and with it, the generation of massive volumes of data, most of which still originates in the radiology department.

Today, more mature forms of artificial intelligence and machine learning can now be brought to bear on a growing repository of imaging studies to improve the detection and treatment of new and existing diseases.

The technology has found a home in oncology and is already revolutionizing cancer care.

“We don’t have enough people to read these images, and we don’t have enough people who can do it to the highest standards,” Constance Lehman, MD, PhD, chief of the breast imaging division at Massachusetts General Hospital and a professor of radiology at Harvard Medical School, told HealthITAnalytics in 2019.

“Deep learning can use full-resolution mammogram images to accurately predict the likelihood of a woman developing breast cancer. Importantly, it is accurate across all races. Existing models are worse than chance at predicting breast cancer in African American women. We need something better than that.”

From academic medical centers (AMCs) to life sciences organizations, researchers are leaning more and more on technology to do the heavy lifting. A significant problem remains having sufficient processing power and data storage to conduct multiple studies, especially concurrently, and reduce the amount of time necessary to turn analysis into insight.

Fortunately, fast and scalable infrastructure is now available to meet the imaging and analytics requirements of cutting-edge studies. Life sciences organizations and AMCs host multiple analytics applications and support a growing number of users. Performance must not come at the expense of the infrastructure’s reliability.

The ability to increase data performance without putting a strain on connected systems — that is, concurrent processing — is the holy grail for research organizations looking to accelerate the detection of disease and the development of therapeutics.

Enterprise imaging can harness the computational and storage power of modern infrastructure solutions by partnering with leading hardware developers in the space, particularly in three areas.

DIGITAL PATHOLOGY

The Digital Pathology Association defines the digital form of pathology as “a dynamic, image-based environment that enables the acquisition, management and interpretation of pathology information generated from a digitized glass slide.”

AMCs and life science organizations are leveraging high throughput scanning and other applications to fuel innovation aimed at reducing lab costs, improving operational efficiency and productivity, and ultimately clinical decision-making and patient outcomes.

By choosing the right technology partner, pathologists can combine imaging knowledge from internal and external sources with infrastructure capable of integrating large volumes of studies and storage of DICOM and non-DICOM images.

MICROSCOPY AND MASS SPECTROMETRY

Data-intensive advanced light and electron microscopy — with 3D microscopy coming into the hold — are generating vast quantities of data requiring a specialized approach for interpreting results. The scanning of individual specimens, for instance, leads to the building of 3D images that require significant data processing and storage needs. The same holds for mass spectrometry imaging.

A modern infrastructure solution can deliver both performance and storage based on the specific needs of individual organizations, big or small.

IMAGE ANALYSIS WITH AI

Artificial intelligence has been around for decades, available in consumer technology in the form of natural language processing. More recently, augmented reality is poised to break through with applications for industry and consumers.

In healthcare, AI has found a strong foothold in radiology. Due to their structure and volume, imaging studies are well suited to machine learning and algorithmic processes. Precision medicine is much closer to reality than fantasy.

At PAIGE.AI, a team of clinical and technological experts is training machines by digitizing millions of histological slides, which generate petabytes of data. The machines are shown thousands upon thousands of examples to recognize specific cancers in new, never-before-seen images. The AI system can then help a pathologist encountering an unknown tissue pattern, with the system quickly scouring digital archives to identify cases with similar morphologies.

The ability to eliminate false diagnoses and make appropriate drugs and therapies available to cure the disease or extend a person’s life comes down to the infrastructure capable of supporting all these processes.

____________________________________

Pure Storage (NYSE: PSTG) gives technologists their time back. Pure delivers a modern data experience that empowers organizations to run their operations as a true, automated, storage as-a-service model seamlessly across multiple clouds. One of the fastest-growing enterprise IT companies in history, Pure helps customers put data to use while reducing the complexity and expense of managing the infrastructure behind it. And with a certified customer satisfaction score in the top one percent of B2B companies, Pure's ever-expanding list of customers are among the happiest in the world.