Radiology is the gateway to finding, assessing, and reporting the course of many diseases. Today radiologists are being asked to quickly and accurately interpret an exploding volume of chest CT and X-ray images. Yet they are using the standard background-impaired imaging interpretation model that includes the interfering normal structures within the chest, like bones and vessels, and machine noise, which can compromise accurate and efficient diagnoses. To think about this another way, this means a radiologist is looking at a “haystack” in the chest and trying to determine—as fast as possible—if there is a needle in it, where it is, and if the needle is growing.
This goal of reducing imaging interpretation turnaround times can impact patient care but also physician wellness. For some healthcare organizations, completing imaging interpretation as quickly as possible can take precedence over meals and resting the eyes and mind, which may result in decreased diagnostic accuracy and increased patient risk.
With the Rise of AI, What IP Disputes in Healthcare Are Likely to Emerge?
Munck Wilson Mandala Partner Greg Howison shared his perspective on some of the legal ramifications around AI, IP, connected devices and the data they generate, in response to emailed questions.
This is compounded by the critical shortage of diagnostic radiologists globally. With 53% of radiologists being over age 55 and imaging procedures increasing by 78% in 10 years, there is a widening gap between supply and demand. It’s part of the reason radiologists are suffering burnout at all-time highs.
Unfortunately, this issue extends well beyond the walls of the Radiology department; it threatens the current healthcare model as imaging is central to nearly all fields of Medicine and Surgery. Deleterious downstream ramifications include increased risk of liability, poor patient outcomes, increased cost of care, and reduced patient satisfaction.
These factors are collectively creating a crisis for both patients and radiologists that must be addressed—now. That is why strategically-oriented healthcare systems are shifting from today’s volume-driven imaging interpretation model to a more predictive, proactive, and preventative one that removes the variables compromising accurate and efficient cardiothoracic diagnoses.
The need for certainty of search
A Personalized Approach to Medication Nonadherence
At the Abarca Forward conference earlier this year, George Van Antwerp, managing director at Deloitte, discussed how social determinants of health and a personalized member experience can improve medication adherence and health outcomes.
The chest is an extremely complex area for imaging interpretation. Finding the needle in the haystack is not easy. Consider that the standard background-impaired imaging interpretation model radiologists use today has significant limitations created by the complex anatomy of the chest. Surrounding structures can conceal disease, making it more difficult to precisely characterize and leading to suboptimal patient care pathways. Radiologists can find it difficult to be sure that a nodule is cancerous or if it is even there because of the impaired view.
Radiologists know that working with standard background impaired X-ray of CT scans, coupled with the exploding volume and complexity of imaging interpretation, is unsustainable. There is an unquestionable need for more certainty of search in visualizing and interpreting images.
Most radiology technology innovation is around triaging imaging reading, but the most valuable advancement may be to just remove the “hay” in the chest with the help of suppression technology.
Suppression technology uses machine learning to remove the variables compromising diagnoses and creates an unimpaired view of the chest. It enables radiologists to precisely detect, characterize, and report findings to improve cardiothoracic diagnostic accuracy and advance earlier detection.
Consider that imaging interpretation with computer-aided detection of lung nodules on CT with a computerized pulmonary vessel suppression function makes a clear difference by reducing missed nodules by 29% and lessening nodule search time by 26%.
Suppression technology yields higher fidelity measurements to ensure optimal treatment plans. It makes it easier to rule out diseases and move on with confidence. It also reduces inter-reader variability and supports holistic analysis and detection of incidental findings.
A clear future for radiology
The current imaging interpretation model undoubtedly has reached its breaking point due to increasing volumes and fewer radiologists. The time is now to solve this systemic imaging interpretation issue.
Patient risk, radiologist burnout, and litigation risk must be minimized to safely increase imaging interpretation loads and gain greater efficiency and accuracy for every read.
The fastest and easiest way to address the challenge is by offering radiologists an unimpaired view of the chest so they can focus on the actionable data in chest imaging.
Being able to find that needle in the haystack with suppression technology is a win for radiologists, patients, health systems, and payers.
Photo: athima tongloom, Getty Images
Dr. Chung is Professor of Radiology, Chief, Section of Thoracic Radiology, and Chief Quality Officer at the University of Chicago Medicine. He specializes in cardiopulmonary imaging and analyzes and interprets chest radiographs and CT scans for thoracic diseases. He also has expertise in interstitial lung disease, occupational lung disease, nontuberculous mycobacterial pneumonia, and diseases of the large and small airways. He has authored more than 180 peer-reviewed articles that have been published in scientific journals and co-authored five book chapters and five books that focus on chest diagnostics.
