Children’s Mercy Kansas City www.childrensmercy.org, Joslin Diabetes Center www.joslin.org and Cambridge Massachusetts-based Cyft Inc. https://www.cyft.com and funded by the Leona M. and Harry B. Helmsley Charitable Trust http://helmsleytrust.org, is going to create a new learning health system to improve the care of type 1 diabetes (T1D).
The funding will enable Cyft to develop the machine learning system that will provide predictive analytics to proactively identify and manage at-risk patients at the point of treatment. Machine learning is an entirely new approach to health analytics because it can generate robust insights from unstructured and imperfect data which may result from the free text notes found throughout EHRs.
Cyft technology will use machine learning and natural language processing as well as device signal processing to analyze multiple data sources and then be able to create predictive models for use by health professionals.
The systems will be deployed at Children’s Mercy and at Joslin since these two organizations have a large number of TID patients. To begin the deployment, Children’s Mercy and Joslin in mid-2017 will deploy machine learning-enabled solutions to proactively manage health outcomes in patients with T1D at two other independent diabetes clinics.
Studies on T1D show that proactive care for glycemic control early in the course of the disease has an influence on long-term clinical outcomes. Managing the disease during childhood is very important as early detection of T1D can make managing the disease more effective. According to Sanjeev Mehta MD, “Predictive analytics holds promise to be able to identify novel clusters of patient factors that could identify high risk patients.”