The CureMetrix Approach
Our CureMetrix team of award-winning physicists and data scientists has a demonstrated competency in quantifying the qualities of anomalies in multi-dimensional, very noisy, signal and image data. We have tested this anomaly quantification in a variety of domains, including population health and genomics.
Today, we are using our expertise to develop the next generation of image analysis technology. Our goal is to assist radiologists with the timely and accurate identification and diagnosis of suspicious anomalies, and to understand the factors that drive these findings.
Developing a Better Way to Triage Anomalies
The CureMetrix approach takes into account the process radiologists use to triage anomalies found in mammograms and other imaging modalities. We are developing our proprietary algorithm to explore pertinent image data, identify what attributes are relevant and discard those that aren’t. The goal is for the algorithm’s predictive powers to improve with each new piece of data processed.
We are leveraging this powerful technology to create a data-driven scoring system, with the goal of enabling radiologists and patients to track breast health from mammogram to mammogram. Training our learning algorithm to create a unique “breast health score” for each image may lead to improved medical image analysis and anomaly quantification that we hope radiologists will be able to rely on to help confirm their evaluation of mammograms and to track their patient’s breast health over time.