Introducing Wellune's new service, breathing pattern similarity analysis for people with dementia.
What is similarity?
The value (unit:%) provided by Welun indicates the similarity between your sleep breathing data and the sleep breathing data of a dementia patient. As of December 2023, Wellune's artificial intelligence model, Dr. Wellune, which has been trained on 6 months of sleep data of 50 patients with severe dementia, compares and analyzes various features between the two data to predict the similarity.
What data do we analyze?
The Wellune sensor collects breathing signal data from an elderly person with mid-stage dementia and breathing signal data from a normal elderly person without dementia symptoms. We select only data with an average signal strength of 10,000 or more per minute. Based on this signal, we extract 'sleep data' that is judged to be sleep by our technology.
How do we analyze it?
A single day's value is actually an average of data over the last 15 days, not a single day. This is done to ensure that the analysis is as accurate and conservative as possible, as daily data can be affected by the conditions and environment of the day. If a user has only been using the service for a short time and does not have enough data for 15 days, we use a random average value. This takes into account the sensitivity of the data, and the model can predict similarity with 99.5% accuracy. The process of taking data (Input) and deriving the final value (Output) is a "black box" and we don't know exactly how it works inside.
Dementia breath similarity, why do you need to know?
This service is important to predict and prepare for future health risks. The numbers provided by Wellune allow users to identify health issues that may arise in the future if current health conditions persist. Based on this, they can improve their lifestyle habits and receive content recommendations that are appropriate for their health condition.