Medical device company, Medtronic, is developing an app that will use IBM Watson’s cognitive computing power to help diabetics and their doctors detect potentially fatal hypoglycaemia events before they occur.
The partnership to integrate IBM’s cognitive computing into Medtronic’s diabetes devices and mHealth platforms was announced at the CES conference last month.
According to IBM and Medtronic officials, the data pulled from blood glucose meters and wearable activity trackers can be analysed by IBM Watson to predict a low blood sugar problem up to three hours before it actually hits. Such insight could enable the patient and their doctor to spot that downward spiral and take steps to prevent it before it becomes a crisis.
“Managing diabetes varies greatly by individual and requires making numerous daily decisions related to eating, activity, exercise and medication in order to keep glucose levels under control,” said President of Medtronic’s Global Diabetes Service and Solutions Unit, Annette Bruls.
“If a person with diabetes glucose levels drops too low, they face the threat of hypoglycaemia, which can cause confusion or disorientation and in its most severe forms loss of consciousness, coma or even death. Conversely, if their glucose level is too high over a long period of time, they risk cardiac disease, blindness, renal failure and amputation of fingers and limbs,” continued Bruls.
IBM and Medtronic have been working on the app since April 2015 and expect to launch it in mid-2016. The algorithms behind the app are based on de-identified data from 150 million patient days gathered through Medtronic’s array of wearables.
The app was used in a pilot with 600 anonymous patient cases. IBM Watson analysed the data from their Medtronic devices to find the indicators of hypoglycaemia and discovered that it could find them hours in advance.
“The models used in this pilot will be designed to continue to evolve according to a user’s nutrition and activity habits along with other contextual information, which may improve the ability of Watson to predict glycaemic excursions and other trends through the Medtronic app,” said Bruls.
“As our collaborative work continues, we expect to make more pattern discoveries that can be built into future additional predictive insights,” concluded Bruls.