Predicting the percentage of people dying with a smartphone How is it possible in 5 years? Research by American researchers shows that this is possible with some tools.
Your smartphone probably vibrates several times a day with different notifications like messages from friends, bank account payments, weather alerts. But in the future, another type of notification may appear on your smartphone, this notification is actually a message from an application that Your risk of death in the coming years predicts.
A new study published in the journal PLOS Digital Health shows that active tracking of movement data through smartphone sensors can effectively predict a person’s risk of death over the next five years with 70 percent accuracy.
This research is based on a large body of evidence showing a link between walking speed and general health. To achieve accurate predictions of health and mortality rates, previous studies generally required participants to use Dedicated fitness trackers 24 hours or completed “gait analysis tests” in laboratory conditions. However, the new study raises the crucial question of whether motion data collected through sensors in conventional smartphones can be sufficient to provide accurate predictions.
Read more: Legal sale of marijuana at US gas stations
Predicting the percentage of people dying with a smartphone
To this end, the researchers examined a large dataset of 100,000 participants from the UK Biobank. This experimental group wore wrist activity monitors for one week and was followed up for at least five years.
This is the largest motion sensor dataset currently available. According to the researchers, the short-term intensity data collected in this series by these fitness trackers is similar to data that can be recorded by a smartphone. After this opinion is possible Predicting the percentage of people dying with a smartphone There is.
“Although these data were collected via motion activity monitors, our sensor models use inputs that can also be collected using today’s low-cost phones,” the researchers explained in the study. “This is possible because of our extensive clinical trials using low-cost phones and the creation of highly accurate predictive models for the health status of heart and lung patients.”
Using only six minutes a day of data to track the intensity of walking, the predictive algorithm can Mortality risk estimates Provide the next 5 years. The accuracy of these data was more sophisticated than estimates made by wearable devices or clinical gait testing; was equal
These findings call for more research with different methods Estimating the mortality risk of people they pay Simple eye scans, blood screening or short balance tests have all been suggested as ways to screen people for early death risk.
Larger experiments are currently planned to focus more on data collected directly from smartphones. Researchers are trying to diversify study groups as much as possible to increase the accuracy of prediction models across different populations.
“This capability is particularly important for health equity goals, given that populations at greatest health risk often have access to the fewest resources,” the researchers concluded in the study. “So people who are more likely to use inexpensive phones rather than expensive wearables will benefit the most from this easy evaluation.”
Smartphone apps can record six consecutive minutes of walking throughout the day, then run predictive models to stratify risk through population analysis of this data.
Read more:
What do you think about predicting the percentage of people dying with a smartphone? Share your thoughts with us in the comments section Technology News With duplicate follow.
Source: newatlas