NO DETERIORATION | These are the 5 patterns of human corruption that we discovered thanks to artificial intelligence

by time news

2024-09-12 07:51:00

Knowledge about the process of aging is still unfair. Understanding something that, in fact, is not uniform or linear, or does not affect everyone in the same way, is very complicated. But those science remains behind the enigma, trying to solve it and confronting it through study.

This is how a group of scientists from the University of Pennsylvania (United States) has identified five ways it is different in that the human brain is damaged with the passing years. A study that has stood out, among other things, for novel method used to get feedback.

AI research

The technique used is known as ‘deep learning‘, an aspect of the Artificial intelligence that rebel to emulate the form of human education to gain knowledge. Thanks to this technology they are able to analyze almost 50,000 brain scansto weigh with the causes of the life -such as cigarettes and alcohol-, as well as genetic and blood markers are linked to them health status and at the risk of disease.

These are the designs

The study, published in the journal ‘Nature’s medicine‘, concludes with some indicators that confirm heterogeneity in the aging brain. Them 5 patterns of brain damage identified are the following:

  1. Subcortical: Your disorder is linked to a set of genes related to stress and pregnancy.
  2. temporal parietal: in this case, it is related to the development of Parkinson’s, Alzheimer’s and schizophrenia.
  3. Lobe for a while: is associated with factors such as birth weight, memory impairment and the progression of dementia.
  4. Cortical diffusion: atrophy of this area can be linked to tobacco and alcohol consumption, diet and multiple sclerosis.
  5. Perisylvian area: related to chronic conditions, including cardiovascular conditions and heart conditions, among others.

#DETERIORATION #patterns #human #corruption #discovered #artificial #intelligence

You may also like

Leave a Comment