Ultrasound-embeddable AI sharp at diagnosing clogged carotid arteries
Testing AI’s ability to detect carotid artery disease on ultrasound, U.K. researchers have found their algorithm achieved 90% accuracy, along with 87% sensitivity and 82% specificity, at the task.
The same algorithm was even better at ruling out the disease, hitting 92% accuracy (with 91% sensitivity and 86% specificity) at identifying normal carotids.
Vascular surgeon Ali Kordzadeh, MBBS, MD, of Anglia Ruskin University and colleagues describe their work in a report published June 10 in Vascular [1].
The team prospectively acquired 156 neck ultrasound scans of patients undergoing duplex ultrasound for suspicion of carotid artery disease.
They had the images interpreted by a network built on convolutional neural network architecture.
Along with the up-or-down results above, the researchers found their AI 94% accurate at detecting carotid arteries that had stenosis of less than 50% plaque.
In sonograms of arteries with between 50% and 70% plaque stenosis, the algorithm had 88% accuracy.
And when evaluating arteries with more than 75% stenosis, the AI had 92% accuracy.
“This study demonstrates the feasibility, applicability and accuracy of artificial intelligence in the detection of carotid artery disease in grayscale static duplex ultrasound images,” Kordzadeh and co-authors write.
More:
This network has the potential to be used as a standalone software or to be embedded in any duplex ultrasound machine. This can enhance carotid artery disease recognition with limited or no vascular experience or serve as a stratification tool for tertiary referral, further imaging and overall management.”
More Coverage of Carotid Artery Imaging:
MRI-based carotid artery measurements improve CVD risk assessment
Choosing Wisely alone not enough to drop low-value imaging of clogged neck arteries
‘Free information’: Radiologists can also diagnose COVID-19 when reading CTA imaging for stroke
Tele-robotics enable long distance carotid ultrasound exams
New noninvasive imaging technique can accurately assess CVD risk
Reference:
- Ali Kordzadeh et al., “Artificial intelligence and duplex ultrasound for detection of carotid artery disease.” Vascular, June 10, 2022. DOI: https://doi.org/10.1177/17085381221107465
Dave P. has worked in journalism, marketing and public relations for more than 30 years, frequently concentrating on hospitals, healthcare technology and Catholic communications. He has also specialized in fundraising communications, ghostwriting for CEOs of local, national and global charities, nonprofits and foundations.