New brain analysis technique discovered by Mount Sinai AI to forecast dementia and Alzheimer's disease | Dil Current News

Brain analysis technique discovered

Overall, the trained deep learning system was able to predict the presence of cognitive impairment "with a modest accuracy that was much greater than chance," according to the creators of the AI model.

AI model was able to identify myelin traits that were related to cognitive impairment. Along with a shrinking protective layer.

The complex mix of self- and caregiver-reported symptoms, a physical examination, and either a PET scan or a spinal tap to look for signs of amyloid plaque buildups in the brain are used currently to diagnose Alzheimer's disease.

However, a novel approach based on artificial intelligence might make the diagnostic procedure considerably more impartial. The AI, which was created by Mount Sinai researchers and presented in a study that was just published this week in the journal Acta Neuropathologica Communications, was trained to detect cognitive impairment with a minimum of human supervision.

RELATED

Is AI the solution to solving the puzzle of Alzheimer's? Imeka, a company in neuroimaging, is betting on it.

The medial temporal lobe and frontal cortex of the brain were examined by the researchers in order to construct the AI. From there, scientists taught a deep learning system to look for changes in the quantity of myelin acting as a protective covering surrounding the brain's nerves by feeding it slide photos of brain autopsy tissues obtained from more than 700 old people.

With the use of the training, an AI model was able to identify myelin traits that were related to cognitive impairment. Along with a shrinking protective layer, this was also accompanied by a non-uniform distribution of myelin throughout the brain tissue, with the majority of the shrinking quantities concentrated in the white matter, the region of the brain associated with learning and neurological function.

Overall, the trained AI was able to identify the existence of cognitive impairment "with a modest accuracy that was significantly greater than chance," according to the study's authors.

The cancer AI startup at Mount Sinai discovers that its algorithms can also detect Parkinson's disease

The researchers expressed optimism that the deep learning model and other comparable algorithms could one day be used to improve diagnosis of Alzheimer's disease and other neurodegenerative conditions. Although the deep learning model may not be ready for widespread use just yet, follow-up analyses were unable to fully understand how the AI reached some of its predictions.

Future study may involve scaling the model to find new paths to improve the screening for cognitive impairment as well as delving into the reasoning behind the AI's judgments to increase their accuracy and dependability.