NVIDIA Inception startups tackle gaps in breast cancer care. Approximately two minutes per breast is now required for a standardized scan using iSono Health's FDA-cleared ATUSA platform, a significant reduction from the up to 45 minutes needed with conventional handheld ultrasound. The company is one of several in the NVIDIA Inception program applying artificial intelligence to address widening gaps in breast cancer care, a field where 40 million mammograms are performed annually in the U.S. amid a projected shortfall of tens of thousands of radiologists. "Getting the scan closer to the patient is the first breakthrough," said Neda Razavi, CEO of iSono Health, envisioning increasingly informative scans to aid clinical decision-making. ATUSA ultrasound streamlines breast screening access. This accelerated acquisition is enabled by artificial intelligence trained on over 1.5 million ultrasound frames, automating image capture with NVIDIA GPU acceleration and open source medical imaging technology. The resulting 3D scan is reportedly 28% more sensitive than traditional 2D handheld ultrasound, offering a potential leap in early detection capabilities. Beyond speed and sensitivity, ATUSA addresses a critical limitation of manual ultrasound: consistency. Handheld techniques are subject to operator variability, making longitudinal comparisons difficult. ATUSA captures each breast the same way every time, creating repeatable views of tissue that facilitate analysis of changes over successive scans. Currently available through partner clinics in California, Texas, Georgia, Tennessee and Washington D.C., the system is expanding to new locations as demand increases. The platform's potential extends beyond initial detection, integrating into a broader effort to streamline the entire continuum of breast cancer care. The company is building a platform that combines imaging exams with pathology results, genomic testing and other biological inputs, and uses AI to create a more comprehensive patient profile. The ultimate goal, according to iSono Health, is to provide clinicians with increasingly informative scans, enabling more informed and personalized therapy decisions. "Our vision is to make that scan increasingly informative: helping clinicians see what is there, understand what has changed and make more informed decisions." Learn AI Tools Whiterabbit.ai reduces radiologist workload with AI density analysis. Whiterabbit.ai's WRDensity software automatically assesses breast density from mammograms, a capability now used in the care of hundreds of thousands of patients. The company's approach directly addresses a critical bottleneck in radiology workflows, where radiologists are increasingly burdened by high volumes of images and a shrinking workforce. Jason Su, cofounder and chief technology officer of Whiterabbit.ai, explained the core challenge: "Every day, breast radiologists face a needle-in-a-haystack problem, trying to find roughly one cancer in every 200 mammograms." The company's AI is designed to help clinicians sift through those images more efficiently, flagging potential areas of concern and reducing the number of false positives. The technology relies on a cluster of NVIDIA GPUs housed at Washington University in St. Louis, with additional GPU capacity used in the cloud for model training. Learn AI Tools Inference, the process of applying the trained AI to new images, runs on NVIDIA GPUs deployed directly within the clinic, enabling rapid analysis at the point of care, the company says. This speed is important, as delays in diagnosis and treatment can significantly impact patient outcomes. Beyond density assessment, Whiterabbit.ai is also developing WRRisk, a clinical decision support software for estimating a patient's long-term risk of developing breast cancer. Su emphasized the potential for AI to augment, rather than replace, the expertise of radiologists. "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," he said, outlining a vision of collaborative care where AI handles routine tasks, allowing radiologists to concentrate on complex cases and critical decision-making. Ataraxis AI predicts chemotherapy response & recurrence risk. Ataraxis AI is developing predictive models that use digital pathology slides to assess chemotherapy effectiveness, potentially reducing the need for invasive tissue biopsies and shortening critical wait times for patients. The company's approach focuses on interpreting patterns within these slides, associating variations with both recurrence risk and a patient's likely response to chemotherapy, a significant advancement over existing tools, according to NVIDIA. Learn AI Tools "The tools oncologists rely on today to guide therapy decisions were largely trained once, fifteen years ago, and never updated," said Joseph Cappadona, member of technical staff at Ataraxis AI, highlighting the static nature of current predictive methods. These AI models integrate digital pathology with standard clinical variables to forecast treatment response and recurrence risk, offering a more complete assessment than previously possible. One model predicts the likelihood of tumor shrinkage before surgery with presurgical chemotherapy, while another estimates five-year recurrence risk and the potential benefit of chemotherapy post-surgery. Treatment decisions often hinge on genomic assays sent off to outside labs, assays that take weeks to process at a moment when speed and certainty matter most. The potential extends beyond individual patient care, as the models improve with each new clinical trial data set acquired. "But as we scale our models, the bigger shift is being able to answer more questions to help oncologists personalize therapy across all cancers," Cappadona explained, emphasizing the broader implications of their work. The computational demands of this analysis are met through NVIDIA technology, providing the processing power needed to analyze hundreds or thousands of images quickly. SimBioSys creates 3D tumor models for precision treatment. SimBioSys is developing a platform integrating diverse patient data to create comprehensive tumor profiles. The company's technology constructs accurate three-dimensional models of breast tumors, including surrounding vasculature and soft tissue, offering clinicians detailed insights for surgical planning and treatment selection. This approach moves beyond traditional two-dimensional imaging by incorporating volumetric data from breast MRI alongside tumor pathology and clinical information. The platform's capabilities extend to predicting the likelihood of breast cancer recurrence. SimBioSys has built a tool that estimates recurrence risk using the 3D volumetric data, combined with pathology and clinical data, providing a more detailed assessment than current methods. The system uses NVIDIA GPUs for processing, using PyTorch accelerated by NVIDIA CUDA, to manage the computational demands of analyzing these complex datasets. NVIDIA technology gives us the computing power to take hundreds or thousands of images, apply our AI and analyze them quickly," Stevens added. Stay current See today's quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.

Wahl previously collaborated with Whiterabbit.ai on an algorithm to help radiologists judge breast density on mammograms to identify people who could benefit from additional or alternative screening.