One of the companies, iSono Health, has developed the FDA-cleared ATUSA platform, a wearable automated 3D ultrasound system. The device can capture a standardized breast scan in about two minutes per breast, compared with as long as 45 minutes for conventional handheld ultrasound.
Its AI was trained on thousands of full-breast scans totaling more than 1.5 million ultrasound frames. NVIDIA says the platform uses GPU acceleration and open source medical imaging technology to generate 3D scans, and the company says the system is 28% more sensitive than handheld 2D ultrasound.
“Getting the scan closer to the patient is the first breakthrough,” said Neda Razavi, CEO of iSono Health. “Our vision is to make that scan increasingly informative: helping clinicians see what is there, understand what has changed and make more informed decisions.”
iSono Health has also developed AI for lesion detection, 3D segmentation and lesion classification. A multicenter study involving 3,200 patients is underway to further evaluate the platform.
Whiterabbit.ai is focusing on mammography. Its FDA-cleared WRDensity software automatically measures breast density from mammograms and has been used in the care of hundreds of thousands of patients.
The company has also built WRRisk, which estimates a patient’s long-term breast cancer risk, and is researching AI that could help radiologists identify more cancers while automatically screening mammograms that are negative.
“Every day, breast radiologists face a needle-in-a-haystack problem, trying to find roughly one cancer in every 200 mammograms,” said Jason Su, cofounder and chief technology officer of Whiterabbit.ai. “We hope AI can be a powerful sidekick to radiologists, helping to clear away the hay so they can focus their expertise where it matters most.”
The company trains its models on NVIDIA GPUs at Washington University in St. Louis and supplements that capacity with cloud infrastructure. Its inference workloads also run on NVIDIA GPUs inside clinics.
Other startups are applying AI after a diagnosis has been made. Ataraxis AI is developing models that analyze digital pathology slides and clinical variables to estimate recurrence risk and predict how patients may respond to treatment.
One of its models estimates whether chemotherapy before surgery is likely to shrink a tumor enough to produce a response. Another evaluates five-year recurrence risk after surgery and estimates the likely benefit of chemotherapy.
“The tools oncologists rely on today to guide therapy decisions were largely trained once, fifteen years ago, and never updated. Our models get stronger every time we acquire more clinical trial data,” said Joseph Cappadona, a member of technical staff at Ataraxis AI. “But as we scale our models, the bigger shift is being able to answer more questions to help oncologists personalize therapy across all cancers.”
Both Ataraxis models have been validated across more than 10 institutions and multiple clinical trials and are in active clinical use. The company runs them on NVIDIA GPUs across on-premises systems, offsite data centers and cloud infrastructure.
SimBioSys is taking a different approach by building AI-powered 3D models of breast tumors, veins and other soft tissue. Its technology is designed to help physicians plan surgeries and evaluate treatment options.
The company has also developed a tool that estimates breast cancer recurrence risk using 3D volumetric MRI data, pathology information and clinical inputs.
“We’re building a platform that now allows us to take multimodal data — imaging exams, pathology results, genomic testing when applicable and other biological inputs — and, using AI, bring that all together,” said SimBioSys CEO Stacey Stevens. “When we do that, it generates new insights beyond what we had from any individual piece.”
SimBioSys uses NVIDIA MONAI along with CUDA-X libraries including cuBLAS and MONAI Deploy, with workloads running on NVIDIA GPUs in the cloud.
“NVIDIA technology gives us the computing power to take hundreds or thousands of images, apply our AI and analyze them quickly,” Stevens said. “That speed matters because patients and physicians need answers quickly. They can’t afford to wait days or weeks.”
Across these companies, NVIDIA is positioning its computing and medical imaging stack as a common infrastructure layer for breast cancer AI. The tools differ in their specific roles, but they are all aimed at shortening delays, supporting clinicians and improving how breast cancer is detected, assessed and treated.
This analysis is based on reporting from the tech buzz & NVIDIA.
Image courtesy of NVIDIA.
This article was generated with AI assistance and reviewed for accuracy and quality.
