Apple watch diagnoses diabetes using a smart neural network known as DeepHeart to recognize people with their diabetes status 85 per cent accurately.
Researchers from the University of California in San Francisco and Cardiogram, an application developer company have determined that collected heart rate data from the smart Apple Watch was more accurate to distinguish people with or without diabetes, as contributing in a research that involves Android Wear users and Apple Watch.
Cardiogram analyzed over 200 million sensor data from 14,011 people using an Android Wear device or Apple Watch with a Cardiogram app and aggregated the data which had probed the activities of step count, heart rate and others.
Traditional process of diabetes detection needed a glucose-sensing system, so that a condition, known as prediabetes goes undiagnosed and unnoticed sometimes. To overcome this problem, the diagnosis can be potentially done using AI-based Cardiogram’s DeepHeart and Apple Watch, which has that potential to keep users aware about the issues and help them suggest a medical professional.
Johnson Hsieh, co-founder of Cardiogram said in a statement that, “Typical deep learning algorithms are data-hungry, requiring millions of labeled examples, but in medicine, each label represents a human life at risk — for example, a person who recently suffered a heart attack or experienced an abnormal heart rhythm.”
Hsieh added that, “To solve this challenge, researchers applied two semi-supervised deep learning techniques (‘unsupervised sequence pretraining’ and ‘weakly-supervised heuristic pretraining’) which made use of both labeled and unlabeled heart rate data to improve accuracy.”