Summary
Artificial intelligence is increasingly being used to support breast cancer care, from analyzing medical images to helping doctors assess cancer risks and choose treatments. Several healthcare technology companies are developing AI-powered tools designed to address delays in screening, shortages of radiologists, and the time required to obtain information needed for treatment decisions.
Companies participating in NVIDIA’s Inception startup program, including iSono Health, Whiterabbit.ai, Ataraxis AI, and SimBioSys, are developing solutions for different stages of breast cancer care. Their technologies aim to help clinicians analyze medical images, assess patient risk, understand tumors, and make more informed treatment decisions.
Although these tools show potential, further clinical validation and appropriate medical oversight remain important for their safe and effective use.
Why Is AI Needed in Breast Cancer Care?
Breast cancer is the most commonly diagnosed cancer among women in the United States. Early detection and timely treatment can make an important difference, but patients and healthcare providers face several challenges.
Many women miss recommended screening appointments because of practical barriers, including limited access to healthcare facilities. At the same time, radiologists must examine large numbers of mammograms, putting pressure on the available workforce.
Delays can also occur after a diagnosis. Some treatment decisions depend on additional laboratory tests, including genomic assays, which may take several weeks to produce results.
AI could help address these challenges by automating parts of medical image analysis, supporting risk assessment, and extracting useful information from existing clinical data.
NVIDIA provides AI computing infrastructure used by several companies working on these technologies.
1. iSono Health Uses AI for Automated Breast Ultrasound
iSono Health is developing technology designed to make breast imaging more standardized and efficient.
Its ATUSA platform is an automated, wearable 3D quantitative ultrasound system. According to the company, it can capture a standardized breast scan in approximately two minutes per breast, compared with up to 45 minutes for conventional handheld ultrasound examinations.
The system uses AI to automate image acquisition and create three-dimensional representations of breast tissue. The company says its AI was trained on thousands of full-breast scans containing more than 1.5 million ultrasound frames.
How could this technology help?
Traditional handheld ultrasound examinations can vary depending on the person performing the scan. Automated scanning aims to make image acquisition more consistent, potentially allowing clinicians to compare images collected during different visits.
iSono Health is also developing AI capabilities for lesion detection, 3D segmentation, and lesion classification. Its longer-term plans include combining information from ultrasound, mammography, MRI, and clinical records.
The company has a multicenter clinical study involving 3,200 patients underway to further evaluate its platform.
These developments could help clinicians obtain more consistent imaging information, although ultrasound does not replace mammography or other examinations when those tests are clinically necessary.
2. Whiterabbit.ai Aims to Improve Mammogram Analysis
Whiterabbit.ai develops AI software intended to support breast cancer screening and risk assessment.
One of its products, WRDensity, is FDA-cleared software that automatically assesses breast density from mammograms. Breast density matters because dense tissue can make some cancers harder to identify on mammography and is also a factor considered in breast cancer risk assessment.
The company has also developed WRRisk, a clinical decision-support tool designed to estimate a patient’s longer-term risk of developing breast cancer.
In addition, Whiterabbit.ai is researching AI systems that could help radiologists identify suspicious findings while automating parts of the screening process for mammograms assessed as negative.
Why does this matter?
Radiologists review large numbers of mammograms, even though only a small proportion reveal cancer. AI assistance could help prioritize examinations requiring closer attention and reduce the workload associated with routine screening.
The potential benefits include faster image review, fewer unnecessary callbacks, and more efficient use of clinical resources.
However, screening AI must be carefully evaluated to ensure that reducing workload does not come at the expense of missed cancers or other diagnostic errors.
3. Ataraxis AI Uses AI to Help Predict Treatment Outcomes
After breast cancer is diagnosed, doctors must determine which treatment is most appropriate for an individual patient.
That decision can depend on factors such as tumor characteristics, the patient’s health, pathology results, and how the cancer is expected to respond to particular therapies.
Ataraxis AI is developing models that analyze digital pathology slides and standard clinical information to predict treatment response and the risk of cancer returning.
According to the company, one model estimates whether a tumor is likely to respond to chemotherapy given before surgery. Another assesses the risk of recurrence over five years and the potential benefit of chemotherapy after surgery.
How could AI help doctors?
Treatment decisions can be difficult because patients with apparently similar diagnoses may respond differently to the same therapy.
By combining information from pathology images and clinical records, AI models may identify patterns that help doctors estimate which treatments are more likely to benefit individual patients.
Ataraxis AI says its models have been validated across more than 10 institutions and multiple clinical trials and are in clinical use.
These tools are intended to support clinical judgment rather than independently prescribe treatment. Their usefulness depends on how accurately their predictions translate into real-world patient outcomes.
4. SimBioSys Creates 3D Models of Breast Tumors
SimBioSys is developing AI-powered precision medicine technology that creates three-dimensional models of breast tumors and surrounding structures.
The company’s technology uses medical imaging and other clinical information to help doctors understand tumor anatomy. These visualizations could support surgical planning and provide additional information when evaluating treatment options.
SimBioSys is also developing tools to estimate breast cancer recurrence risk using information from breast MRI, tumor pathology, and clinical data.
Why are 3D tumor models useful?
Medical images provide important information, but understanding a tumor’s size, shape, and relationship with nearby structures can be challenging.
Three-dimensional models can help clinicians visualize these relationships more clearly. When combined with pathology and other available patient information, AI may also help identify patterns that are not apparent from a single source of data.
SimBioSys is working toward a multimodal approach that brings together different types of medical information to support more personalized care.
The technology is designed to assist clinical decision-making, and its value depends on appropriate validation and integration into medical practice.
How AI Could Improve the Breast Cancer Care Journey
The technologies developed by these companies focus on different stages of care, from screening to treatment planning.
- creening and imaging: Automated ultrasound and mammogram analysis could help healthcare professionals process medical images more efficiently.
- Risk assessment: AI tools may help estimate the likelihood of developing breast cancer or experiencing a recurrence.
- Diagnosis support: Image analysis and lesion characterization could help clinicians identify areas that require further investigation.
- Treatment planning: Models that predict treatment response may provide additional information when doctors compare treatment options.
- Surgical planning: Three-dimensional tumor visualizations could help clinicians understand tumor anatomy.
The broader goal is to make better use of the medical information already available while reducing some of the delays and workload involved in cancer care.
What Are the Limitations of AI in Breast Cancer Care?
Despite its potential, AI is not a replacement for trained healthcare professionals or established diagnostic procedures.
AI systems can produce incorrect predictions, perform differently across patient populations, or struggle when used in settings that differ from their training data. Medical tools therefore require rigorous testing and ongoing monitoring.
Patient privacy is another important consideration because these systems may process sensitive medical images and clinical records.
It is also necessary to distinguish between a tool that has received regulatory clearance for a specific use and a technology that is still undergoing clinical research. Evidence supporting one application does not automatically establish that the same system is safe or effective for other purposes.
Doctors must continue to interpret AI-generated information alongside a patient’s medical history, examination results, preferences, and other relevant evidence.
What Comes Next?
Healthcare technology companies are working to combine imaging, pathology, genomic information, and clinical records to provide a more complete picture of each patient’s condition.
Further clinical studies will help determine how reliably these tools perform, whether they improve treatment decisions, and whether their use leads to better outcomes for patients.
Wider adoption will also depend on regulatory requirements, cost, integration with hospital systems, and whether healthcare providers can use the technology effectively in everyday practice.
If these challenges are addressed, AI could become a useful addition to breast cancer screening and personalized treatment planning.
Conclusion
AI is being developed to address several challenges in breast cancer care, including imaging workloads, risk assessment, and the time required to make treatment decisions.
Companies such as iSono Health, Whiterabbit.ai, Ataraxis AI, and SimBioSys are exploring different approaches, from automated 3D ultrasound to mammogram analysis, treatment prediction, and tumor visualization.
These technologies do not eliminate the need for established medical examinations or professional judgment. However, with appropriate validation and clinical oversight, they could help healthcare teams use medical information more effectively and support more informed decisions throughout the breast cancer care journey.