As outlined in a recent report published by Academic Radiology, the emergence of “Big Data” analytics could be a potential game-changer for radiology.
“By virtue of an existing robust information technology infrastructure and years of archived digital data, radiology departments are particularly well positioned to take advantage of emerging Big Data techniques,” Akash P. Kansagra, MD, Mallinckrodt Institute of Radiology, Washington University School of Medicine in St. Louis, and colleagues wrote.
These are four specific ways Big Data could change the specialty forever, according to the authors:
1. Make image interpretation more personal
Providing personalized care is a key part of the radiologist’s job. As the authors explained, management of an ovarian cyst in one patient with no risk factors for cancer is much different than management of an identical cyst in a patient with numerous risk factors.
But there is more information in the electronic medical record (EMR) than any individual radiologist can manage on their own.
“Although radiologists already account for some relevant patient-specific factors (e.g., patient age, history of cancer) when interpreting studies, the vast quantity of lesion-, patient-, and population-specific data contained in the EMR exceeds the ability of a radiologist to meaningfully incorporate into interpretation,” the authors wrote. “For example, biopsy samples from a wide range of tumor types now routinely undergo detailed genetic analysis to extract information about DNA, RNA, and protein expression in tumor cells that profoundly influence prognosis and subsequent management.”
Big Data helps radiologists dive deeper into the EMR. It helps them “see” more of the bigger picture, connecting dots that otherwise would have never been noticed.
2. Discovery of new imaging markers
The authors wrote that, in some cases, the brevity that makes a radiologist so effective at his or her job may actually be keeping key information from being seen. Big Data can help identify important imaging features that have not been discovered yet.
“Big Data methods are ideally suited to the challenge of identifying new imaging markers,” the authors wrote. “By virtue of being able to handle extremely large data sets, Big Data methods are likely to outperform conventional, single-institution research efforts with comparatively small sample sizes that may not provide sufficient statistical power to identify rare markers or those with small effect sizes.”
3. More accurately quantify the value of radiology services
Radiologists are being asked demonstrate the value and importance of what they do now more than ever, and Big Data’s data mining ability can work wonders in this regard.
“The true value of a test is represented by the incremental outcome and cost benefit relative to no imaging test,” the authors wrote. “In this context, ‘no imaging test’ may signify empirical treatment without an imaging-supported diagnosis or empirical nontreatment. With this definition, the value of an imaging test may differ dramatically between patients.”
This represents a significant step in finally grasping the actual “worth” of any given test.
4. Improve radiology workflow
The imaging community is always looking for ways to improve quality, and even the smallest change to day-to-day radiology workflow can make a noticeable impact on patient care. With Big Data, that workflow can be analyzed in record time and with remarkable detail.
“Individual institutions can track a variety of operational data related to efficiency, including variations in imaging volumes, utilization rates of specific scanners at different times of the day, technologist-specific scanner throughput, and radiologist-specific report turnaround times,” the authors wrote. “These insights can produce significant improvements in departmental efficiency.”
The full report from Kansagra et al. includes much more information about Big Data, and it can be read in full here.