- Radiology
How AI Can Support Radiologists
MRI and CT – these methods are widely used when it comes to making a diagnosis, be it for injuries, cancer, but also for diseases of the heart, the central nervous system or the skeleton. And they are more needed than ever: A growing and aging population as well as changing lifestyles are contributing to an increase in chronic diseases such as cardiovascular diseases and cancer, leading to a rising number of diagnostic imaging procedures. Time pressure and workload for radiologists are mounting up. Artificial Intelligence (AI) can support radiology to provide accurate and timely diagnosis for patients.n field]
What Challenges Are Radiologists Facing Today?
An estimated 286 million contrast-enhanced X-Ray/CT and MRI procedures are performed annually worldwide1 – and the demand for medical imaging to detect diseases, guide treatment decisions and support therapy planning is growing further, increasing pressure on radiologists. A particular US institution reported that their average radiologist reading CT or MRI examinations would need to interpret one image every 3-4 seconds in an 8-hour workday to meet workload demands.2 In addition, there is a shortage of trained radiologists.3 These trends contribute to the prevalence of burn-out among radiologists.4 As a result, errors may occur. A publication from 2018 speaks of approximately 40 million diagnostic errors involving imaging worldwide every year5.
What Are the Benefits of AI in Medical Imaging for Patients?
What does that mean in practice and where is the benefit of AI in radiology for patients? For example, AI can aid clinicians in the detection of lung nodules in asymptomatic patients. AI triage applications can flag suspected pathologies for prioritized reading of cases that may need urgent diagnosis and treatment, e.g. potential intracerebral hemorrhage (ICH) or large vessel occlusions (LVO). ICH and LVO can be associated with stroke.
Sources
1 Source: Bayer estimates based on various data sources
2 McDonald RJ, Schwartz KM, Eckel LJ, Diehn FE, Hunt CH, Bartholmai BJ, Erickson BJ, Kallmes DF. The effects of changes in utilization and technological advancements of cross-sectional imaging on radiologist workload. Acad Radiol. 2015 Sep;22(9):1191-8. https://pubmed.ncbi.nlm.nih.gov/26210525/
3 The Complexities of Physician Supply and Demand: Projections From 2018 to 2033. AAMC. https://www.aamc.org/media/10886/download#:~:text=Demand%20for%20physicians%20continues%20to,demand%20will%20grow%20more%20steeply.&text=By%202033%2C%20demand%20for%20physicians,full%2D%20time%2Dequivalent%20physicians.; the Royal colleague fo Radiology: https://www.rcr.ac.uk/system/files/publication/field_publication_files/clinical-radiology-uk-workforce-census-2020-report.pdf
4 Chetlen AL, Chan TL, Ballard DH, Frigini LA, Hildebrand A, Kim S, Brian JM, Krupinski EA, Ganeshan D. Addressing Burnout in Radiologists. Acad Radiol. 2019 Apr;26(4):526-533. https://pubmed.ncbi.nlm.nih.gov/30711406/; Physician Burnout in Radiology: Perspectives From the Field, Cheri L. Canon et al. (2022) https://pubmed.ncbi.nlm.nih.gov/34494444/
5 Fundamentals of Diagnostic Error in Imaging, Itri et al. (2018) https://pubs.rsna.org/doi/full/10.1148/rg.2018180021