Advances in AI — April 2024

Emphasizing multi-structure segmentation and feature extraction from chest CT scans, an emerging AI model demonstrated an approximately 70 percent AUC for predicting significant incidental extrapulmonary findings as well as two-year and 10-year all-cause mortality.

In the third episode of a three-part podcast, Anand Narayan, M.D., Ph.D., and Amy Patel, M.D., discuss the challenges of expanded breast cancer screening amid a backdrop of radiologist shortages and ever-increasing volume on radiology worklists.

Providing automated TI-RADS classifications and worksheets, the new AI-enabled software may facilitate improved efficiency with thyroid ultrasound exams.

In the second episode of a three-part podcast, Anand Narayan, M.D., Ph.D., and Amy Patel, M.D., discuss recent studies published by the Journal of the American Medical Association (JAMA) that suggested moving to more of a risk-adapted model for mammography screening.

While noting the strength of MRI for complete staging of disease and ultrasound’s ability to provide local disease characterization, the authors of a new literature review suggest the two modalities offer comparable results for diagnosing pelvic endometriosis.

Diffusion-weighted MRI provided pooled sensitivity and specificity rates of 82 percent and 81 percent respectively for gauging patient response to concurrent chemoradiotherapy for esophageal cancer, according to new meta-analysis.