As the annual Radiological Society of North America (RSNA) conference approaches, one thing is certain: AI for radiology will remain a central topic. For ultrasound stakeholders seeking to navigate the buzz and bring home actionable insights to ignite their AI transformation, a focused agenda is critical. It’s important to differentiate between future innovation and the AI tools already transforming radiology workflows and patient care. Here are some key AI-powered solutions currently available that you should explore at the RSNA 2024.
Scan smartly with ultrasound AI
The quality of ultrasound images often depends on the operator's skill and experience. Radiologists rely on detailed, high-quality images for accurate diagnoses, but an operator's training, technique, and adherence to protocols may affect the consistency and reliability of the images. In some workflows, radiologists must review and approve images before patients are dismissed or the exam is added to the reading worklist. Inconsistent or inadequate images can cause delays, require repeat exams, or lead to patient recalls, all of which may impact workflow efficiency and the patient experience.
AI-supported image acquisition can empower ultrasound operators of all experience levels to capture high-quality, standardized diagnostic images of the heart. Currently available with GE HealthCare’s Vscan AirTM and VenueTM family of ultrasound equipment, Caption GuidanceTM offers real-time, step-by-step probe guidance, to assist with precise probe positioning. By enabling users without prior ultrasound scanning experience to perform scans, this advanced technology can help expand access to care.
A key feature of Caption Guidance is its quality meter, which actively assesses the image as it is being captured. As the image improves and approaches diagnostic quality, the meter rises, providing immediate feedback to the user. Once optimal conditions are met, the system automatically captures the image without the need for manual intervention.
Save time to curb burnout
AI solutions can help combat radiologist burnout and boreout by alleviating the tedium of some repetitive and routine tasks associated with image interpretation. For instance, radiologists often spend considerable time manually measuring complex anatomical structures and documenting their findings. Ultrasound AI algorithms can detect and highlight anatomical structures and pathologies, while providing measurements of their size, shape, and other key characteristics. These AI-generated measurements can be seamlessly integrated into reports, eliminating the need for manual data entry.
As an example, the LOGIQ™ Auto Abdominal Imaging Suite 1.0, with features like Auto Preset Assistant, Auto Abdominal Color Assistant, and Auto Renal Measure Assistant, standardizes image acquisition, ensuring consistent, high-quality abdominal scans. This improves radiologists' efficiency and allows them to focus on interpretation rather than technical issues. The Auto Renal Measure Assistant for LOGIQ automatically detects the kidney and measures the length, height, and width, driving efficiency. Adding AI-driven lesion detection in dense breast tissue via QVCAD™ on Invenia™ ABUS 2.0 can reduce reading time by 33%1, with up to 93%2 sensitivity for lesion detection.
Support decisions with AI-enhanced insights
AI tools enhance radiologists' diagnostic capabilities by automating image analysis, detecting subtle patterns, and providing quantitative assessments across various imaging modalities. They provide a plethora of advantages such as improved diagnostic accuracy, workflow efficiency, and personalized patient care.3
For example, GE HealthCare’s LOGIQ platform and Invenia ABUS 2.0, powered by Koios DS™ technology, offers tools like Breast Assistant that support breast lesion assessment. AI compares each patient's lesion against a database of more than 900,000 images, helping clinicians detect disease earlier and reducing unnecessary procedures on non-malignant lesions. Studies have shown that Breast Assistant can help provide results in two seconds* and reduce benign biopsies by 55%.4, 5 Similarly, the Thyroid Assistant on LOGIQ provides real-time AI-driven thyroid cancer risk assessments, automatically populating TI-RADS® descriptors, and creates a Koios AI adapter risk assessment based on AI models, helping to further increase clinical confidence and reducing inter-reader variability by 41%.*
Test your AI IQ at RSNA 2024
Unlock ultrasound AI with GE HealthCare’s VerisoundTM AI & Digital solutions. Our innovations are transforming ultrasound departments around the world. Learn how by attending an expert-led, AI-focused session in our Ultrasound Precision Care Education Room in booth 8343 in the North Hall. Space is limited, so reserve your spot today to secure your seat at these exclusive events.
Be sure to also visit GE in booth 7330 in the North Hall to explore our AI-powered ultrasound features firsthand. While you’re there play the Verisound AI Bowl—compete for the top score on our leaderboard.
Request a meeting or demo at RSNA 2024
REFERENCES:
1.) Yulei Jiang, “Interpretation Time Using a Concurrent-Read Computer-Aided Detection System for Automated Breast Ultrasound in Breast Cancer Screening of Women with Dense Breast Tissue,” American Journal of Roentgenology 211, no. 2 (2018): 452–461.
2.) Victoria L. Mango et al., “Should we ignore, follow, or biopsy? Impact of artificial intelligence decision support on breast ultrasound lesion assessment,” American Journal of Roentgenology 214, no. 6 (2020): 1445-1452.
3.) Reabal Najjar, "Redefining Radiology: A Review of Artificial Intelligence Integration in Medical Imaging," Diagnostics (Basel) 13, no. 17, (2023):2760.
4.) Lev Barinov et al., “Impact of Data Presentation on Physician Performance Utilizing Artificial Intelligence-Based Computer-Aided Diagnosis and Decision Support Systems,” Journal of Imaging Informatics in Medicine 32, no. 3 (2018): 408–416.
5.) Lev Barinov et al., “Decision quality support in diagnostic breast ultrasound through artificial intelligence,” IEEE Signal Processing in Medicine and Biology Symposium (SPMB), 2016.
*Koios Medical internal data. Available upon request. Result availability is dependent on the available network bandwidth.
Caption Guidance, Invenia, LOGIQ, Venue, Verisound, and Vscan are trademarks of GE HealthCare. QVCAD is a trademark of Qview Medical, Inc. Koios DS is a trademark of Koios Medical. BI-RADS ATLAS and TI-RADS are trademarks of the American College of Radiology.
Product and features may not be available in all countries and regions. Full product technical specification is available upon request. Contact a GE HealthCare representative for more information.
Venue family consists of Venue, Venue Go, Venue Fit, and Venue Sprint.
GE is a trademark of the General Electric Company, used under trademark license.
JB31737XX
