Artificial intelligence (AI), if it hasn’t already, is on the brink of transforming many industries, and based on the significant increase in the number of FDA-approved AI medical devices and algorithms in recent years, this trend is finding a foothold in healthcare too. While embracing artificial intelligence within physician workflows and clinical pathways is a learning curve, it also presents an opportunity to “superpower the data,” driving value in care for patients, physicians and hospitals/health systems, particularly to meet the population health challenges of today.
Among the many uses and types of AI technology in a clinical setting, the subset of augmented intelligence, which incorporates both machine learning and powerful predictive modeling, offers important capabilities that can enhance detection of subclinical disease and accelerate early intervention of chronic conditions, at scale.
Radiologists have the opportunity to stand at the forefront of this advancement. With a vast volume of imaging data and an ever-increasing demand for imaging, there’s a wealth of big data to be mined. At the same time, short supply of radiologists calls for a need to go beyond the standard read and extract valuable and actionable clinical insights – all while being user-friendly. It may seem like a tall order, but AI technology, when embedded in computed tomography (CT) scans and other routine imaging equipment already in use, improves report quality to radiologists to make opportunistic diagnoses of subclinical conditions or incidental findings that may have otherwise gone undetected. This can be a matter of life and death for individuals with conditions like osteoporosis and coronary artery disease, which often don’t present clinical symptoms in early stages, but pose severe health concerns down the line.
From a value-based care perspective, disease prevention and early intervention measures help health systems avoid future costs associated with complications and catastrophic events stemming from undetected and unaddressed chronic disease. On top of the profound quality of life impact for patients with chronic conditions, the economic costs are staggering, accounting for 90% of the $3.5 trillion healthcare expenditure in the U.S.[i]
Nanox.AI has developed a tool that is geared entirely toward detecting findings which correlate to chronic conditions such as coronary artery disease — CAD — (based on elevated levels of coronary artery calcification) and osteoporosis (based on vertebral body compression fractures and low bone mineral density), and providing physicians with actionable insights to help direct early intervention known to help prevent catastrophic cardiac events, such as heart attack, and stroke, as well as hip and other major osteoporotic fractures down the line.
Such secondary evaluation of the population, at scale, can reduce the overall cost of care for the health system; in this way, it delivers value of mutual benefit to various stakeholders to make an impact across the board.
To illustrate this, osteoporosis is associated with hip and other major fractures, often resulting in hospitalization and an increased 12-month mortality rate of up to 25% for those with hip fractures.[ii] Direct spending on non-traumatic fractures is estimated to be $25.5 billion in 2025.[iii] Cardiovascular disease, the leading cause of death worldwide, contributes over $225 billion to U.S. health spending.[iv],[v] The devastating impact of these two chronic conditions underscores the need for health systems to shift toward a proactive, preventative approach to care, through improved early detection of subclinical disease and creation of robust clinical pathways to treat these diseases early.
Presenting new technologies to existing workflows and processes presents challenges in any industry and is especially true in the notoriously slow-to-adopt healthcare field. And this is not without good reason; practitioners are overburdened with increasing numbers of patients and cumbersome electronic health record (EHR) requirements, and hospitals and health systems often lack the operational capacity and/or infrastructure to utilize AI technology to its full capacity. However, because augmented AI integrates seamlessly into existing radiology technology, the additional capabilities are minimally burdensome for patients and providers alike. One study evaluating Nanox.AI within a U.S. health system demonstrated that the addition of this augmented AI technology added a negligible amount of time – to review and verify a series of imaging – to the standard workflow.[vi]
Looking beyond the challenges of adopting new technologies, embedded AI technology augments the standard read of medical imaging by leveraging clinical recommendations that physicians can tailor to the unique needs of their patients. New solutions for cardiac and bone health are just the beginning, but they demonstrate how technological improvements in medical imaging can help enable early intervention to shape the trajectory of a patient’s care journey.
As AI technology advances with ever growing applications in the healthcare sector, health systems and practitioners have the opportunity to embrace a new, highly efficient “partner” in medicine at earlier stages in the clinical pathway. While AI is only part of the equation, it is primed and ready to help hospitals and health systems build the right clinical pathways, drive value, and, ultimately, improve patient care.
[i] CDC Centers for Disease Control and Prevention
[ii] S Schnell, SM Friedman, DA Mendelson, KW Bingham, SL Kates. The 1-year mortality of patients treated in a hip fracture program for elders. Geriatr Orthop Surg Rehabil. 2010 Sep;1(1):6-14. doi: 10.1177/2151458510378105. 0
[iii] R Burge, B Dawson-Hughes, DH Solomon et al. Incidence and economic burden of osteoporosis-related fractures in the United States, 2005-2025. J Bone Miner Res. 2007 Mar;22(3):465-75.doi: 10.1359/jbmr.061113.
[iv] CW Tsao, AW Aday, ZI Almarzooq, A Alonso, AZ Beaton, MS Bittencourt, et al. Heart disease and stroke statistics – 2022 up-date: a report from the American Heart Association. Circulation 2022 Feb 22; 145(8): e153–639.
[v] D De Smedt, K Kotseva, D De Bacquer, D Wood, G De Backer, J Dallongeville, et al. Cost-effectiveness of optimizing prevention in patients with coronary heart disease: the EUROASPIRE III health economics project. Eur Heart J 2012 Nov 1; 33(22): 2865–72. doi: 10.1093/eurheartj/ehs210.
[vi] A Kurek, D Langholz, A Ahmed. Closing the Loop in AI, EMR, and Provider Partnerships: The Key to Improved Population Health Management? Telehealth and Telemedicine Today. 2022