Among the most powerful tools that enable modern day healthcare delivery are advanced imaging techniques. Often, these tools are used to validate physical exams, visualize internal anatomy, or triangulate pathology based on subjective symptoms.

A recently published study in the journal Healthcare found that in the last decade alone, the rates of CT and ultrasound examinations have nearly doubled, indicating a growing trend in the use of imaging in healthcare delivery. There are many potential reasons for this, including the fact that medical imaging techniques have become incredibly advanced, meaning that clinicians can use these tools to determine the exact problem and navigate treatment swiftly. Accordingly, there is a significant amount of investment and growing interest in making these tools intelligent, easy to use, and more accessible.

Organizations are quickly embracing this call-to-action. Last week, devices and services giant GE, which has a prominent footprint in healthcare, announced that it was awarded a $44 million grant from the Bill & Melinda Gates foundation to develop AI assisted ultrasound technology. The press release discusses how one of the primary goals behind this initiative is to create a more user-friendly interface that will enable clinicians to better support a wide variety of healthcare screening techniques, with a specific eye to improving healthcare outcomes and access in low to middle income countries.

Roland Rott, President and CEO of Ultrasound at GE Healthcare, explains that although ultrasound technology is an incredibly powerful tool for screening and diagnostics in the field, “a key limitation is the guidance of lesser-skilled users to effectively apply affordable point-of-care ultrasound in their care environment.” The hope behind this grant will be to help bridge this gap and guide users to capture better images and diagnostic information with the device.

This conceptual drive to improve imaging techniques is growing across the spectrum of different modalities.

Earlier this year, a study was published in the journal Radiology, which indicated that a highly specialized AI algorithm was able to predict breast cancer risk with notable levels of accuracy. The impact of this study is momentous; although experts have been discussing the use of AI tools in clinical medicine for years, examples like this provide very real scenarios in how this can actually be embedded in a clinician’s workflow.

Moreover, these tools also have great potential in being used directly by patients. Take for example DermAssist, a tool developed by Google Health to help users find information about their skin concerns. As the service describes it: users can “upload 3 photos of your skin condition and answer a few questions. Using what it has learned from millions of skin-related images, DermAssist then looks for signs of various skin conditions in your submitted photos and information.” Though still in its preliminary stages, tools like this have incredible potential to change the way people approach healthcare.

These are just a few examples of how imaging modalities are slowly inculcating the leading work being done in the worlds of artificial intelligence and machine learning. Of course, the technology still has a long way to go and requires further development. However, if done correctly, these tools have immense potential. Although a clinician has the unique ability to take into account other considerations such as patient context, a longitudinal view of the patient, and social factors, AI tools like these may still be of value in being able to triage cases or act as clinical decision-making support systems. Indeed, as imaging has become such an integral part of the practice of modern medicine, this field of innovation can truly impact the world of healthcare delivery.