The Role of AI in Ultrasound
Artificial Intelligence (AI) is having a banner year in 2023. With the advent of generative AI in particular, innovation is on an exponential upswing. Nowhere is this more evident than in healthcare, where AI offers promising insights. Clinicians need to look only to ultrasound to see the impact AI is having on healthcare, from assessment and diagnosis to treatment and recovery.
New technology enhances a proven modality
Ultrasound as a technology is going through its own period of innovation right now. The introduction of point-of-care ultrasound (PoCUS) has reduced the need for traditionally bulky diagnostic imaging equipment, while capacitive micromachined ultrasonic transducers (CMUTs) have markedly enhanced the quality of ultrasound scans.
Along with those advances, ultrasound has greatly benefitted from AI, including:
- AI can automate diagnostic assessments through image analysis, measurement, and reporting. This can free sonographers to focus on other aspects of patient care, while also speeding ultrasound reviews and results.
- Machine learning can increase diagnostic accuracy by identifying patterns and features in ultrasound images that are often difficult for humans to see. This is particularly useful in the diagnosis of complex conditions, including certain types of cancer.
- AI is being used to analyze ultrasound images in conjunction with other data sources, such as patient history and laboratory results. This helps clinicians identify risk factors for disease, track patients’ progression, and improve critical decision-making.
It’s important to recognize that, in many cases, AI use cases are siloed. As AI-assisted ultrasound becomes more commonplace in practice, clinicians will have powerful automated workflows at their fingertips and will be better able to leverage ultrasound quickly and effectively for diagnosis and treatment.
AI benefits patients and providers
Considering the enormous strides generative AI has taken in 2023, technologists are on the cusp of even more innovative ultrasound applications. That has benefits for patients and providers alike.
For patients, the improved accuracy of imaging can offer more certainty in their diagnosis and confidence in their prescribed treatment plan. It also opens the door to personalized care at a lower cost, thanks to decisive action on the part of providers. And, with fewer rescans required, patients save time, money, and anxiety.
For providers, the ability to diagnose and begin treatment quickly paves the path to better patient prognoses. Moreover, AI makes it easier to train clinicians on both ultrasound equipment usage and scan interpretation, giving them confidence in their ultrasound-based diagnoses and interventions.
Put simply, AI and ultrasound effectively bridge the gap between patient conditions and provider interventions.
Prominent AI and ultrasound applications
As ultrasound has become a focal point in healthcare, so too have its AI applications. Clinicians from emergency rooms to obstetrics wards are extoling the virtues of AI-assisted solutions and the critical insight ultrasound provides. Take, for example,
- Cardiology, where AI is being used to diagnose the heart, assess the severity of cardiovascular conditions, and guide the placement of pacemakers and defibrillators.
- Oncology, where AI can detect and characterize tumors, assess the response to treatment, and predict the risk of recurrence based on scans and dataset analysis.
- Obstetrics and gynecology, where AI has the power to automatically detect fetal anomalies, measure fetal growth, and predict preterm birth.
What’s next for AI and ultrasound?
That’s just the beginning. As the technology advances, we can expect to see even more use cases and practical opportunities to combine AI and ultrasound to improve the diagnosis and management of disease.
In fact, ultrasound manufacturers are already integrating AI into their technology in meaningful new ways. GE’s LOGIQ™ series, for example, is designed with AI-assisted technology to help diagnosticians improve ultrasound scan quality. And the Food and Drug Administration (FDA) recently granted 510(k) clearance for Sonio Detect, a fetal ultrasound software as a service (SaaS) platform that uses AI to enhance the assessment of fetal heart and brain structures.
Up next for AI and ultrasound? There’s already speculation that AI-powered ultrasound machines will soon perform exams, identify abnormalities, and provide diagnoses instantly with the same accuracy as human diagnosticians — but with no need for a sonographer.
While there’s no telling what the future holds, all signs point to powerful AI/ultrasound combinations. If so, AI-enhanced ultrasound is likely to be game-changing for healthcare and the patients we serve.