Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×
AI Medicine Science

AI Researchers Predict Alzheimer's Years Before Diagnosis (sciencedaily.com) 43

Slashdot reader pgmrdlm quotes Science Daily: Timely diagnosis of Alzheimer's disease is extremely important, as treatments and interventions are more effective early in the course of the disease. However, early diagnosis has proven to be challenging. Research has linked the disease process to changes in metabolism, as shown by glucose uptake in certain regions of the brain, but these changes can be difficult to recognize... Researchers trained [a] deep learning algorithm on a special imaging technology known as 18-F-fluorodeoxyglucose positron emission tomography (FDG-PET). In an FDG-PET scan, FDG, a radioactive glucose compound, is injected into the blood. PET scans can then measure the uptake of FDG in brain cells, an indicator of metabolic activity.

The researchers had access to data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), a major multi-site study focused on clinical trials to improve prevention and treatment of this disease. The ADNI dataset included more than 2,100 FDG-PET brain images from 1,002 patients. Researchers trained the deep learning algorithm on 90 percent of the dataset and then tested it on the remaining 10 percent of the dataset. Through deep learning, the algorithm was able to teach itself metabolic patterns that corresponded to Alzheimer's disease. Finally, the researchers tested the algorithm on an independent set of 40 imaging exams from 40 patients that it had never studied. The algorithm achieved 100 percent sensitivity at detecting the disease an average of more than six years prior to the final diagnosis.

"We were very pleased with the algorithm's performance," Dr. Sohn said. "It was able to predict every single case that advanced to Alzheimer's disease

This discussion has been archived. No new comments can be posted.

AI Researchers Predict Alzheimer's Years Before Diagnosis

Comments Filter:
  • by Anonymous Coward

    What do they tell the patient? "Good news! You won't start losing your marbles until six years from now?"

    • by ShanghaiBill ( 739463 ) on Saturday November 10, 2018 @01:57PM (#57622440)

      What do they tell the patient? "Good news! You won't start losing your marbles until six years from now?"

      1. There are lifestyle changes that can delay the onset of dementia. Changes in the diet, more exercise, more social activity.

      2. There is a lot of promising research, including immunotherapy, and cellular senescence. If potential dementia patients can be detected early, then this research can be directed more efficiently, and cures found sooner.

      3. People can make better life plans with more information. They can update their will, set up a trust, spend more time with their grandchildren, take that long planned vacation to Outer Mongolia, or whatever.

      • Changes in the diet

        In particular, a low-carb diet.

        • In particular, a low-carb diet.

          Yes, eat less carbs, especially sugar, and less saturated fat. Eat more fish, and monounsaturated fat (olive oil, peanut oil). All of these are negatively correlated with dementia.

          Lose weight. Exercise more. Get out of the house and do stuff. Vary your routine.

          • I don't worry about saturated fats. You can't really avoid them, especially not if you're already avoiding carbs, and picking mainly natural whole foods. Our ancestors have eaten saturated fats for millions of years, and our own body makes saturated fat to store excess energy.

            Stuff like peanut oil should not be a major part of your calorie intake. Also, you can't trust that oils are really clean and pure and not oxidized. For example, Italian police claims that half of the Italian "extra virgin" olive oi

            • You can taste the difference and you can see the difference in some cases if you refrigerate it.

              Some olive oils are extremely rich and tangy. You wouldn't normally use them for cooking.

              I've had "pure" olive oil products which wouldn't solidify in the refrigerator. That's a bad sign.

            • Stuff like peanut oil should not be a major part of your calorie intake.

              Why not? What's wrong with peanut oil?

              For example, Italian police claims that half of the Italian "extra virgin" olive oil has been mixed with cheap oil.

              I have heard that about half of "Italian olive oil" is actually from Spain.

    • For one, if this is accurate in the sense that there were no false positives, then it could be used to see how effective preventative treatments are as they can then determine people who do not have Alzheimer yet but would normally develop it. They can also look at the difference in why some people developed it 2 years later, and others 6 years later. If some treatment delays it by 10 more years then that's fantastic, and they can begin working on methods to better trace the progression, perhaps realize dif

  • The results? (Score:4, Interesting)

    by Ol Olsoc ( 1175323 ) on Saturday November 10, 2018 @01:56PM (#57622436)
    Everyone get tested, so that people who come down with Alzheimer's will be called a pre-existing condition.
  • IF it's really 100% (or close) accurate, how soon can it be used for the general populace?

    Alzheimer's costs a LOT to deal with. Beyond avoiding some suckage for the patient and especially their family, this could save some serious money if universally applied.

    It's kinda invasive with requiring an injected marker, but that's already done pretty routinely for kidney trouble etc.

  • by Ungrounded Lightning ( 62228 ) on Saturday November 10, 2018 @02:01PM (#57622448) Journal

    The algorithm achieved 100 percent sensitivity at detecting the disease an average of more than six years prior to the final diagnosis.

    So no false negatives. But how many false positives? TFA doesn't say.

    You can get 100% detection of fires by sounding the alarm continuously. But that's not very useful. Without that other number we don't know if this is useful or bull generated fertilizer.

    • Re: (Score:3, Informative)

      So no false negatives. But how many false positives? TFA doesn't say.

      The false positive rate is 18%.

      So if the test says you don't have it, you don't have it. If it says you do have it, there is a 1 in 5 chance that you don't.

      Here is the original paper [rsna.org]

      An obvious next step is to train the NN on a lot more data.

      • by Anonymous Coward

        No no no no no, that's not how statistics work.

        If the false positive rate is 18%, and 1% of people will get Alzheimer's, and it says you have it, there is approximately a 1 in 18 chance that you have it.

        • If the false positive rate is 18%, and 25% of people will get Alzheimer's, and it says you have it, there is approximately a 25 in 18 chance that you have it.

          I don't think so.

      • If it says you do have it, there is a 1 in 5 chance that you don't.

        No, you have to consider the base rate. Of all adults over 65 in the US, about 10% have Alzheimer's.

        So of 100 people, 10 will get Alzheimer's. Since the test's sensitivity appears to be perfect, all 10 will get a positive if tested, along with 18 others who won't get Alzheimer's. That's a total of 28 positive results. So while 18 of 100 results are false positives, 18/28 = 64% of positive results are false.

        So, if this test says you are going to get Alzheimer's, there's a 36% chance that you actually

  • But it forgot to tell us.

  • for a loooong time already. Also do not buy this vendor lock in pseudo security, not built-into hard-drives, or this Apple T2 chip thing. Can't review, don't know what it does, can't trust. Use open source reviewable stuff that we can also fix in case of an issue. I for one only trust open solutions an non of the pseudo "military grade" fluff that breaks as soon as someone looks too close, https://www.youtube.com/watch?... [youtube.com]
  • Talk about sensitivity (how few false negatives) is important, but you also need high specificity - i.e. how few false positives.

    In this case, their specificity was only 82% of the time, that is, a significant issue.

    Human doctors were less likely in general to say you had the disease, with only a sensitivity of about 57% but their greater reluctance to say "You are mentally impaired) meant they had a 91% specificity.

    Personally, I would not go to a doctor that told one in five healthy people they were sick.

    L

"The vast majority of successful major crimes against property are perpetrated by individuals abusing positions of trust." -- Lawrence Dalzell

Working...