*ISLAMABAD – A new research suggests that now AI might also be able topredict premature death in people and that too very accurately.*
Experts from University of Nottingham recently trained an AI system toevaluate a decade of general health data submitted by over half a millionpeople in the UK. The AI was then tasked with predicting if individualswere at risk of dying prematurely from chronic disease.
The predictions of early death made by the AI algorithm were ‘significantlymore accurate’ than the predictions delivered by a model that did not usemachine learning, said study’s lead author Stephen Weng, as per *LiveScience.*
In order to evaluate the patients’ premature death, the team tested twotypes of AI: ‘deep learning’ AI in which layered information-processingnetworks help a computer to learn from examples, and a simpler ‘randomforest’ AI that combines multiple, tree-like models to consider possibleoutcomes.
*Artificial Intelligence outshines doctors at predicting heart diseasedeathslink*
Then, the team compared the AI models’ conclusions to results from astandard algorithm, known as the Cox model. Based on these three models,the team evaluated the database of genetic, physical and health datasubmitted by over 500,000 people between 2006 and 2016. During that time,around 14,500 of the participants died, primarily from heart disease,cancer, and respiratory diseases.
Read Top Stories
All the three models determined that factors like age, gender, smokinghistory and prior cancer diagnosis were top variables for assessing thelikeliness of a person’s early death. But, the researchers found that themodels diverged over other key factors.
Where the Cox model leaned greatly on ethnicity and physical activity, themachine-learning models did not. The random forest model focused more onbody fat percentage, waist circumference, amount of fruit and vegetablespeople ate, and skin tone, whereas the deep-learning model emphasized onjob-related hazards and air pollution, alcohol intake and the use ofcertain medicines.
After everything, the deep-learning algorithm gave the most accuratepredictions, accurately identifying 76% of subjects who died during thestudy period. The random forest model, in comparison, predicted about 64%of premature deaths, whereas the Cox model identified only 44%.
The study co-author Joe Kai said that machine learning can be used tosuccessfully predict mortality outcomes over time. Using AI this way ‘couldhelp with scientific verification and future development of this excitingfield’.
