This startup aims to help surgeons detect breast cancer with 3D imaging

Revolutionizing breast cancer surgery with AI-powered 3D visualization: SimBioSys, an Illinois-based startup, has developed TumorSight, an innovative technology that transforms routine MRI images into detailed, color-coded 3D models of breast tumors and surrounding tissue, aiming to improve surgical planning and outcomes for breast cancer patients.

Key features of TumorSight:

• Converts black-and-white MRI images into spatially accurate, volumetric 3D visualizations
• Uses distinct colors to highlight different breast structures (e.g., red for veins, blue for tumors, gray for surrounding tissue)
• Allows surgeons to manipulate the 3D model on a computer screen for better insights
• Calculates crucial measurements such as tumor volume and distance from chest wall and nipple
• Provides data on tumor-to-breast volume ratio to aid in surgical decision-making

Potential impact on breast cancer treatment:

• Nearly 2.3 million women are diagnosed with breast cancer annually worldwide
• Breast cancer claims over 500,000 women’s lives each year
• Approximately 100,000 women in the U.S. undergo some form of mastectomy annually
• TumorSight aims to improve surgical planning and potentially reduce unnecessary mastectomies

Advantages over traditional methods:

• Offers a significant improvement compared to standard radiology reports and limited images
• Eliminates the need for surgeons to consult radiologists for additional information
• Aims to standardize presurgical imaging and move surgical planning from “art to science”

Technical infrastructure:

• Utilizes NVIDIA A100 Tensor Core GPUs in the cloud for model pretraining
• Employs NVIDIA MONAI for training and validation data
• Leverages NVIDIA CUDA-X libraries, including cuBLAS and MONAI Deploy
• SimBioSys is a member of the NVIDIA Inception program for startups

Additional AI innovations in development:

• Conversion of prone MRI images to realistic 3D visualizations of upright patient positioning
• Accounts for gravity’s impact on breast tissue and skin elasticity
• Rapid risk analysis for cancer recurrence based on tumor features, pathology reports, and patient demographics

Implications for cancer recurrence prevention:

• Current molecular analysis of removed tumors can take up to six weeks
• SimBioSys’s AI model aims to generate risk analysis within hours
• Potentially matches or exceeds traditional risk of recurrence scoring methods
• Offers significant time and cost savings compared to conventional approaches

Broader context: As AI continues to advance in healthcare, technologies like TumorSight represent a growing trend of using machine learning and advanced visualization techniques to enhance surgical planning and improve patient outcomes. These innovations have the potential to significantly impact breast cancer treatment strategies, potentially reducing the need for invasive procedures and improving long-term survival rates.

Future implications: While TumorSight and related technologies show promise in improving breast cancer surgery outcomes, their long-term impact on patient care and survival rates remains to be seen. As these AI-powered tools become more widely adopted, it will be crucial to monitor their performance in real-world clinical settings and assess their ability to consistently improve surgical decision-making and patient outcomes across diverse populations.