Artificial intelligence system capable of detecting Alzheimer's disease years before conventional...

Artificial intelligence system capable of detecting Alzheimer's disease years before conventional... https://i0.wp.com/www.eresviral.com/wp-content/uploads/2018/11/Sistema-de-inteligencia-artificial-capaz-de-detectar-el-Mal-de-Alzheimer-años-antes-del-diagnóstico-convencional.jpg?fit=160%2C146&ssl=1

Artificial intelligence system capable of detecting Alzheimer's disease years before conventional diagnosis





Artificial intelligence system capable of detecting Alzheimer's disease years before conventional diagnosis


A timely diagnosis of Alzheimer's disease is extremely important, given that treatments and other interventions are more effective in the early part of the course of the disease. However, achieving an early diagnosis is very difficult. Various investigations have linked the disease process to changes in metabolism, such as those seen in the absorption of glucose in certain regions of the brain, but these changes can be difficult to recognize.



The team of Jae Ho Sohn, Benjamin Franc and Yiming Ding, of the University of California in San Francisco (UCSF), United States, decided to test whether through an artificial intelligence system prepared by them it would be possible to find changes in brain metabolism that predict Alzheimer's disease. For this, they resorted to the strategy of machine learning, a type of artificial intelligence that allows computer programs to learn things when they are exposed to new data, without needing to be reprogrammed.



The researchers prepared a deep learning algorithm (a modality of machine learning) and trained it in the analysis of images obtained using a special technique of positron emission tomography. In this technique, a radioactive glucose compound is injected into the blood and then the scan makes measurements indicative of the level of metabolic activity in brain cells.







Brain images obtained from scans with a special technique of positron emission tomography. Image A corresponds to a 76-year-old man with Alzheimer's disease. B, to an 83-year-old woman with mild cognitive impairment. The C, a man of 80 years without Alzheimer's disease or mild cognitive impairment. The patient with Alzheimer's had a little less gray matter in his brain than the patient without Alzheimer's disease or mild cognitive impairment. The difference between the patient with mild cognitive impairment and the one who did not have Alzheimer's disease or mild cognitive impairment was practically imperceptible to the naked eye. (Images: Radiological Society of North America)



The researchers trained the deep learning algorithm with 90 percent of a set of more than 2,000 images of that type taken from a thousand patients and then tested it with the remaining 10 percent of the database, in order to check how much I had learned. Through deep learning, the algorithm was able to teach itself the metabolic patterns that correspond to Alzheimer's disease.



Finally, the researchers tested the algorithm on an independent set of 40 scans of as many patients as it had never studied. The algorithm reached a 100 percent sensitivity in the detection of the disease, and not only that: it did so using images with an advance of approximately six years regarding the definitive diagnosis.



The results are very promising, and now it will only be necessary for them to validate themselves by testing the system with a much greater number of cases and with the pertinent verification by various independent institutions and the competent official organisms.



Additional Information

        .

  !function(f,b,e,v,n,t,s)

  {if(f.fbq)return;n=f.fbq=function(){n.callMethod?

  n.callMethod.apply(n,arguments):n.queue.push(arguments)};

  if(!f._fbq)f._fbq=n;n.push=n;n.loaded=!0;n.version='2.0';

  n.queue=[];t=b.createElement(e);t.async=!0;

  t.src=v;s=b.getElementsByTagName(e)[0];

  s.parentNode.insertBefore(t,s)}(window, document,'script',

  'https://connect.facebook.net/en_US/fbevents.js');

  fbq('init', '369524843414444');

  fbq('track', 'PageView');

    .


SOURCE LINK ERESVIRAL.COM 
https://www.beviral.online