Artificial intelligence is increasingly being used in healthcare, and breast cancer is becoming an important area where the technology could help doctors make faster and more informed decisions.
Several healthcare startups are now developing AI systems that can assist with different stages of breast cancer care, from breast imaging and cancer-risk assessment to treatment planning.
Many of these companies are part of NVIDIA’s Inception startup program and are using AI computing technologies to process medical images and clinical information.
Summary
AI is being developed to support breast cancer care in several areas, including medical imaging, risk prediction and treatment decisions.
Companies such as iSono Health, Whiterabbit.ai, Ataraxis AI and SimBioSys are developing different approaches that could help clinicians analyze breast scans, identify potential cancers, estimate risks and understand how patients may respond to treatment.
The technology is not intended to replace doctors. Instead, the goal is to provide clinicians with additional information and reduce some of the workload involved in breast cancer diagnosis and treatment.
What Happened?
Breast cancer remains one of the most commonly diagnosed cancers among women in the United States. At the same time, the healthcare system faces several challenges, including missed screenings, growing demand for medical imaging and a shortage of radiologists.
Around 40 million mammograms are performed in the U.S. each year. With the number of medical images increasing, radiologists are under pressure to examine more scans.
Another challenge comes after a cancer diagnosis. Treatment decisions can depend on pathology and genomic testing, some of which may take weeks to complete.
AI companies are now working on tools that could help address some of these delays.
AI Could Improve Breast Imaging
iSono Health is developing an automated 3D ultrasound system called ATUSA.
The company says the system can capture a standardized 3D image of the breast in about two minutes per breast. This compares with conventional handheld ultrasound examinations, which can take considerably longer.
One of the potential advantages of automated imaging is consistency.
With handheld ultrasound, the quality and coverage of an examination can depend partly on the person operating the ultrasound probe. An automated system can capture the breast in a more standardized way, potentially making it easier to compare scans taken at different times.
iSono Health says its AI technology was trained using thousands of full-breast scans containing more than 1.5 million ultrasound frames.
The company is also developing AI capabilities for detecting lesions, creating 3D segmentations and classifying lesions.
A multicenter clinical study involving about 3,200 patients is underway to further evaluate the platform.
AI Could Help Radiologists Find Cancer
Another startup, Whiterabbit.ai, is developing AI technology for breast cancer screening.
Its WRDensity software automatically evaluates breast density from mammograms and has already been used in clinical settings.
The company is also developing WRRisk, which is designed to help estimate a patient’s long-term risk of developing breast cancer.
Whiterabbit.ai is researching additional AI systems that could help radiologists identify suspicious findings while automatically handling mammograms that appear negative.
The idea is not to replace radiologists. Instead, AI could act as another layer of support, allowing specialists to spend more time on cases that require closer examination.
AI Could Help Predict Treatment Response
Finding cancer is only one part of the process. After diagnosis, doctors also need to determine which treatment is most likely to work for a particular patient.
This is an area where Ataraxis AI is developing AI-based clinical tools.
The company’s systems analyze digital pathology slides along with other clinical information to predict treatment response and recurrence risk.
One of its models is designed to predict whether chemotherapy given before surgery is likely to significantly reduce the size of a patient’s tumor.
Another model estimates the risk of cancer returning within five years and the potential benefit of chemotherapy following surgery.
According to the company, these models have been evaluated across multiple healthcare institutions and clinical trials.
If such systems continue to prove accurate in clinical practice, they could potentially provide doctors with additional information when deciding between treatment options.
3D Models Could Give Doctors a Better View of Tumors
SimBioSys is taking another approach by combining medical imaging and other patient information to create three-dimensional views of breast tumors and surrounding tissue.
Its technology can create 3D models using information from medical imaging and other clinical sources.
The company is also developing technology designed to estimate breast cancer recurrence risk using information such as breast MRI data, pathology and clinical information.
The broader goal is to combine different types of medical data rather than examining each piece separately.
For example, imaging may show the physical characteristics of a tumor, while pathology can provide information about the tumor’s biological characteristics. Combining those sources could potentially give doctors a more complete picture.
Why It Matters
Breast cancer care involves many stages, and delays or limitations at any one stage can affect the overall patient experience.
AI could potentially help in several areas:
- Make medical imaging more consistent
- Help radiologists analyze large numbers of scans
- Identify suspicious areas that may require further examination
- Estimate a patient’s cancer risk
- Predict how a tumor may respond to treatment
- Combine imaging, pathology and clinical information
- Provide doctors with additional information when planning treatment
However, these technologies still require careful clinical validation.
An AI system used in healthcare needs to be accurate, reliable and properly evaluated before it can become part of routine patient care.
What Comes Next?
The next stage for these technologies will be broader clinical testing and validation.
Companies will need to demonstrate that their systems work reliably across different hospitals, patient populations and clinical situations.
Regulatory approval will also remain an important part of the process. Some of the technologies discussed are still investigational and may not be approved for commercial medical use.
As medical AI develops, the focus is likely to move beyond simply detecting abnormalities. The bigger opportunity could be combining imaging, pathology, genomic information and patient history to provide doctors with a more complete picture of each patient’s condition.
Conclusion
AI is beginning to play a role in different parts of the breast cancer care journey.
From automated ultrasound and mammogram analysis to treatment-response prediction and 3D tumor visualization, startups are exploring how AI can support doctors and improve the way medical information is processed.
The technology is not a replacement for medical professionals. Instead, its most useful role may be helping clinicians handle large amounts of information, identify important findings and make better-informed decisions.
The effectiveness of these systems will ultimately depend on clinical evidence, regulatory approval and how safely they can be integrated into real-world healthcare.