Being able to anticipate or predict something way before it happens allows us to better prepare ourselves. In the case of our health, sometimes an early diagnosis means that there could be a chance of you curing the problem entirely, or at least preventing it from affecting you too badly and making adjustments to your life accordingly.


This is where the use of AI in the medical field seems to play a big role in doing, and according to researchers at UC San Francisco, it appears that they have managed to develop an AI system that can actually detect Alzheimer’s in a patient as early as six years before the actual diagnosis has been made.

According to Jae Ho Sohn, one of the researchers behind the project, “One of the difficulties with Alzheimer’s disease is that by the time all the clinical symptoms manifest and we can make a definitive diagnosis, too many neurons have died, making it essentially irreversible.” This is why the AI was developed to help detect the disease in its extremely early stages.

To train the AI, the researchers fed it PET scans from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). This dataset contained scans from patients who were eventually diagnosed with Alzheimer’s, to mild cognitive impairment, or no disorder. It seems that eventually the system managed to learn on its own which features were deemed to be important when it came to diagnosing the disease.

The initial tests proved to be pretty successful, but more testing and calibration still needs to be done, so it might be a while before we see it start being used in hospitals.

Filed in Medical. Read more about , and .

Related Articles on Ubergizmo