Zebra Medical Vision, a machine and deep learning startup utilizing big data to diagnose diseases far better than radiology has received CE approval for its 7th disease detecting algorithm that will detect one of the deadliest of all: breast cancer. The algorithm is capable of detecting suspected malignant lesions in mammography scans. It is the latest addition to other automated tools announced in the past as part of it’s “All-In-One” AI1 business model, among them algorithms that automatically detect brain bleeds, vertebral fractures, coronary artery disease, osteoporosis and more.With its preparatory technology, Zebra-Med helps doctors and millions of their patients receive fast, accurate medical image analysis nearly instantly.
According to the American Cancer Society, breast cancer makes up 25% of all new cancer diagnoses in women globally – with nearly 1.7 million women being diagnosed annually, and statistically has a lot of ‘miss’ or ‘false positive’ cases.
Existing software solutions, called Mammo CAD (computer aided-detection) have been marketed for a number of years – attempting to assist mammographers in identifying suspicious lesions in mammography scans. Unfortunately, the large number of false alarms, coupled with a price tag that has placed these products within reach of only wealthier healthcare economies, have not led to widespread adoption globally.
Zebra’s Mammography algorithm aims to change that dynamic, by providing a state of the art malignancy detection product at a previously unprecedented price point. The first version to be released supports 2D Hologic devices, and Zebra expects to add support for additional vendors, as well as 3D support during the course of 2019. The algorithm broadens Zebra-Med’s AI1 “All-In-One” Imaging Analytics package, which has already analyzed more than 1M scans in over 5 countries.
“Early detection of breast cancer is a crucial component of disease prevention,” says Dr. Michael Fishman, a breast imaging radiologist at Beth Israel Deaconess Medical Center in Boston, Massachusetts. “An accurate AI assistant can provide a significant boost to radiologists seeking to provide the best care for their patients by increasing detection and limiting false positives.”
