This is the first in Ellumen’s new series on AI Innovation in Medical Imaging
Artificial Intelligence or AI technology, a broader term that often encapsulates Machine Learning (ML), Deep Learning (DL), Neural Networks (NN) and Robotic Process Automation (RPA), is ubiquitous in our lives today. Whether you are asking Alexa what the weather is like outside, depositing a check on your mobile banking app or taking an Uber home from the office, you are leveraging AI technology. As consumers, we are reaping the benefits of AI daily in the form of a more personalized user experience, more efficient service and ultimately, greater customer satisfaction.
As healthcare systems across the world strive to improve clinical outcomes and increase efficiency, AI stands out as a technology with the potential to revolutionize the healthcare dynamic. The value of collecting, analyzing and ultimately driving clinical decisions based on a robust data set is not a novel concept. In fact, this method is universally recognized as best practice by clinicians and healthcare administrators around the world. As a tool to support clinicians, AI offers the prospect of simultaneously enhancing scale of data input while improving the speed and efficiency of analysis and decision making. This capability is particularly advantageous to clinical disciplines with extremely large and complex data sets like medical imaging.
A recent study from Definitive Healthcare found that over half (53%) of responding organizations (hospitals and medical imaging centers) say that they will be utilizing AI within the next two years to either perform or assist in patient imaging or business operations. The application of AI in medical imaging has already proven its ability to increase productivity, reduce errors, improve diagnostic accuracy, enhance predictive analysis and reduce expenditures:
A Stanford study utilized an AI algorithm to read chest X-rays for 14 different pathologies. Findings showed the algorithm matched expert radiologists in accuracy while reducing read time (reducing average read time from 4 hours with radiologists to less than 2 minutes with AI).
Google has developed an AI system that can grade prostate cancer cells with 70% accuracy, improving on benchmarks for human pathologists (which averaged 61% accuracy).
A 2018 study published in Radiology found AI able to detect Alzheimer’s disease in brain scans 6 years prior to diagnosis with 98% accuracy by analysis of glucose levels difficult to perceive by human specialists.
These are just a few examples of the many promising applications for AI technology in the field of medical imaging. So, with so much potential and tangible benefit, why are some organizations reticent to incorporate AI technology? The most common explanations provided for slow or nonexistent AI adoption in the study performed by Definitive Healthcare were high costs, lack of strategic direction and lack of technical expertise. These all represent legitimate concerns, but the ability to overcome these challenges is well within reach.
Ellumen is excited to explore the benefits of AI technology as a tool to assist our partners in the medical imaging community. AI holds the potential to improve the experience for patients and clinicians, enhance clinical care and ultimately improve healthcare outcomes.. Leveraging our experience supporting the medical imaging community at the Department of Veterans Affairs, National Cancer Institute National Biomedical Imaging Archive (NBIA) and Department of Defense, we will explore AI innovation in a cost-efficient, impactful and clinically-accessible manner. Be sure to stay tuned for Ellumen’s upcoming blog series on the delivery of AI innovation in medical imaging, and feel free to reach out to me for more information on how you can unlock the potential of AI technology for your business.