Imaging: Artificial intelligence may improve interpretation of MRI

Contrast-enhanced ultrasound for detecting renal cell carcinoma recurrence following ablation and new protocols for low-dose computed tomography in stone patients were among the other research highlights in imaging at the AUA annual meeting. The take-homes were presented by Simpa Salami, MD, of the University of Michigan, Ann Arbor.

• Multiple studies examined the use of artificial intelligence in MRI. The technology can help determine who needs an MRI, improve the interpretation of MRI, and facilitate prostate cancer localization.
• A novel method to measure bladder wall micromotion with M-mode ultrasound was validated in pigs.
• Contrast-enhanced ultrasound was used to detect renal cell carcinoma recurrence following ablation.
• New protocols were offered for low-dose computed tomography (CT) in stone disease.
• Surgeons risk greater exposure to radiation when they are relatively short in stature or stand relatively farther from the source of radiation. A separate study found that lead glasses can reduce the amount of radiation absorbed by the eyes, but still allow up to 56% to pass through.
• Fluorodeoxyglucose positron emission tomography/computed tomography was found to have high specificity but low sensitivity for staging pelvic lymph nodes in patients with penile carcinoma.
• Prostate-specific membrane antigen (PSMA) PET/magnetic resonance imaging improved the detection of prostate cancer recurrence.
• Multiple studies showed that membranous urethral length predicts urinary continence recovery, and one study showed that perineal ultrasound assessment of pelvic floor function provides accurate assessment of membranous urethral length.
• In a retrospective cohort, 52% of false-positive lesions on MRI fusion prostate needle biopsy were due to inflammation.
• A pig study suggested a step forward for image-guided surgery with a new small-molecule intravenous nerve-binding fluorophore.