por
John R. Fischer, Senior Reporter | May 11, 2022
iSono Health's 3D breast ultrasound system
iSono Health has designed and received FDA clearance for the first automated and wearable 3D breast ultrasound system.
Attached to a wearable accessory, the ATUSA solution is designed to make 3D breast ultrasound accessible at the point of care as well as make patients feel comfortable when being scanned. The compact device comes with intuitive software for automated image acquisition and analysis, and can scan the entire breast volume in just two minutes without an experienced operator. The software allows for images to be displayed in real time, provides 3D visualization and localization of tissue and seamlessly integrates advanced machine learning models that help clinicians make decisions and manage patients.
iSono Health's co-founder and CEO Maryam Ziaei told HCB News that the portable and automated solution, combined with machine learning, would be valuable worldwide, particularly in countries with limited resources. Ultrasound has been shown to be an effective alternative to mammography for breast cancer diagnosis in developing countries, especially among women with dense breasts. “The ATUSA system was designed to improve access to millions of women worldwide. The other important application of deep learning models, especially in low-resource settings, is to triage patients that require additional diagnostic imaging, since there is limited access to radiologists and imaging centers.”
Ad Statistics
Times Displayed: 40734
Times Visited: 137 GE HealthCare’s Repair Center Solutions are an ideal complement to your in-house service team. We service a broad range of mobile devices, including monitors and cardiology devices, parts, and portable ultrasound systems and probes.
ATUSA provides reproducible imaging and performs longitudinal monitoring to detect changes in breast tissue based on the patient’s baseline. It can be used to monitor high-risk patients, benign lesion follow-ups and responses to treatment for personalized breast care.
Compared to the solution, conventional ultrasound is not as specific and results in more false positives. And for handheld systems, image quality is highly dependent on the skill of the operator, creating a lack in standardized interpretation, according to Ziaei. “The addition of deep learning models to breast ultrasound can significantly improve specificity and reduce false positives and false negatives. Automation will provide consistent image quality independent of operator skill, and in turn will reduce the variation in interpretation of the images and unnecessary biopsies and therefore, improve patient experience.”
The company plans to submit more solutions to the FDA for clearance, and is currently performing prospective case collection studies to test various deep learning models integrated with ATUSA for breast lesion localization and classification.
Back to HCB News