Now we may prepare for our stress beforehand!
A new study by a group of scientists from Columbia University may help tackle chronic stress. Researchers have created a neural network that predicts the stress level of a person with an accuracy of 85% based on the analysis of recorded parameters and feedback.
How the study was conducted?
Throughout the year, scientists monitored the activity of 79 citizens of New York using fitness bracelets. The smart devices took into account the intensity and regularity of training, and also recorded the physical parameters of the body. Twice a day, participants received an SMS message with questions on the bracelet:
For a preliminary calculation of the stress level “for tomorrow”, the data and responses for the last three days were analysed. If the participant forgot to answer the questions, the data for all three days were not taken into account. Additionally, the length of daylight hours, air temperature and precipitation were recorded – this data was taken from the report of the city meteorological station.
The data obtained helped to reveal interesting patterns:
Predicting the level of stress makes it possible to intervene promptly and prevent its progress. For example, if a doctor knows that inclement weather drives the patient into stress, and, according to the forecast for tomorrow it will be raining and the usual jogging will be cancelled, he can give appropriate recommendations to stabilize the condition.
Perhaps soon, technology and smart devices will help us prevent stress at the household level too.