#nvidiaa100

SimBioSys

A startup uses AI-powered 3D visualizations to assist surgeons in identifying breast cancers.

A potential new technique to assist surgeons in operating on and better treating breast malignancies is a new AI-powered imaging-based system that generates precise three-dimensional models of tumors, veins, and other soft tissue.

SimBioSys is a business that enhances cancer treatment via biophysical simulations and artificial intelligence (AI):

Technology

With the use of SimBioSys‘s technology, clinicians may better comprehend a patient’s condition and tailor therapy by creating virtual models of tumors.

How SimBioSys Works

The system developed by SimBioSys creates a risk analysis for a patient’s cancer by examining their tumor, pathology report, and demographic information. This aids physicians in promptly identifying the most effective course of action to prevent recurrence.

Goals

By giving doctors a deeper grasp of each patient’s condition, SimBioSys hopes to revolutionize cancer treatment decision-making.

Partners

Partners Reputable organizations including the University of Chicago Medicine, Cedars Sinai, Cleveland Clinic, City of Hope, and Mayo Clinic are among those with whom SimBioSys has worked.

In 2018, Tushar Pandey, John Cole, and Joe Peterson created SimBioSys. The corporate office is located in Chicago, Illinois.

The technique, developed by the Illinois-based firm SimBioSys, transforms standard black-and-white MRI pictures into volumetric, spatially precise pictures of a patient’s breasts. After that, it shows various sections of the breast in different hues. Tumors are depicted in blue, surrounding tissue is gray, and the vascular system, or veins, may be red.

Using 3D Visualizations

The 3D visualizations may then be readily manipulated by surgeons on a computer screen, giving them valuable information to assist direct procedures and inform treatment strategies. TumorSight is a device that computes important surgical metrics, such as the volume of a tumor and its distance from the nipple and chest wall.

Additionally, it offers important information on the extent of a tumor relative to the total volume of the breast, which may assist surgeons in deciding before a treatment starts whether to attempt breast preservation or opt for a mastectomy, which sometimes has unpleasant and cosmetic side effects. TumorSight was approved by the FDA last year.

WHO estimates 2.3 million women worldwide are diagnosed with breast cancer yearly. Over 500,000 people die from breast cancer yearly. According to Brigham and Women’s Hospital, 100,000 US women get mastectomy yearly.

Surgeons get a radiological report with one or two tumor pictures and the phrases, “Tumor size and location.” In order to get further information, the surgeon must locate the radiologist, speak with them (which is not always possible), and go over the case with them.

SimBioSys pretrains its models using cloud-based NVIDIA A100 Tensor Core GPUs. Additionally, it runs its imaging technology using NVIDIA CUDA-X libraries, such as cuBLAS and MONAI Deploy, and leverages NVIDIA MONAI for training and validation data.

SimBioSys is a participant in the NVIDIA Inception startup program

SimBioSys is already developing other AI applications that it believes will increase the survival rate of breast cancer patients.

It has created a brand-new method for resolving breast MRI pictures collected while the patient is face down and transforming them into realistic, virtual 3D visualizations of the tumor and surrounding tissue that will be seen when the patient is face up during surgery.

Surgeons will find this 3D visualizations particularly useful as it allows them to see how a breast and any tumors would appear after surgery.

The technique computes how gravity affects various types of breast tissue and takes into consideration how different types of skin elasticity affect a breast’s form while a patient is on the operating table in order to produce this images.

The business is also developing a novel approach that uses AI to swiftly provide insights that may prevent cancer from returning.

Today, pathology tests are performed in hospital laboratories on malignancies removed by surgeons. After that, the samples are submitted to another outside lab for a more thorough molecular study.

Usually, this procedure takes up to six weeks. Patients and physicians are unable to promptly develop treatment strategies to prevent recurrence if they do not know the aggressiveness of the cancer in the resected tumor or how that kind of cancer would react to various therapies.

The new technology from SimBioSys analyzes a patient’s demographic information, the hospital’s first tumor pathology report, and the 3D volumetric properties of the recently excised tumor using an AI model. Based on such data, SimBioSys creates a risk analysis for that patient’s cancer in a few hours, assisting physicians in making an immediate decision on the best course of action to prevent recurrence.