WHITE PAPER
Artificial Intelligence in Ultrasound Imaging
This white paper provides a practical over view of artificial intelligence (AI) in ultrasound, focusing on real-world applications for image acquisition, analysis, workflow, and clinical decision support. It is designed for sonographers, radiologists, and other imaging professionals who want to understand how AI is currently used, where it adds value, and what the limitations are.
Smarter imaging
Stronger insight
Better decisions
From Algorithm to Clinical Support
Machine Learning
Algorithms that learn from data to identify patterns and make predictions,
Deep Learning
Neural networks that analyze complex image data and enable high-performance tasks such as recognition and segmentation.
Generative AI
Models that can create new content, assist with image interpretation, and support education and reporting.
Clinical Decision Support
AI tools that integrate imaging findings with clinical information to support (not replace) professional judgment.
Where AI Is Used In Ultrasound
AI is being applied across the ultrasound workflow, from image acquisition to clinical interpretation.
Acquisition Guidance
Real-time guidance to help obtain standard views and improve image quality.
View Recognition
Automatic identification and classification of anatomical views.
Segmentation
Automatic delineation of organs, structures, and lesions.
Automated Measurements
AI-enabled measurements of structures such as cardiac chambers, fetal biometry, and more.
Doppler & Quantification
Automated Doppler analysis and quantitative blood flow assessment.
Decision Support
Integration of imaging findings with clinical data to assist with interpretation and management.
Foundation/Vision-Language Models
Large, multi-modal models that can integrate images and text, with potential for broader ultrasound applications.
​
Generative AI
Creation of synthetic images, reporting assistance, education tools, and new approaches to image enhancement and simulation.
Quantitative Ultrasound
AI-driven extraction of quantitative imaging biomarkers - for example, tissue characterization, radiomics, and risk stratification.
​
AI- Assisted CEUS
Automated analysis of contrast-enhanced ultrasound for lesion detection, characterization, and treatment response assessment.
Emerging Directions
New areas of AI research and development are expanding the potential of ultrasound.