AI now makes it possible to detect several diseases early. Some forms of cancer are already better identified in patients with Al and soon could be the order of Alzheimer's disease.
For Alzheimer's disease, it is well known that early diagnosis can better support patients and provide them with more effective treatments to slow the progression of the disease. A study published in the American Journal of Radiology explains that the onset of Alzheimer's disease coincides with a series of metabolic changes including glucose uptake in certain areas of the brain. Unfortunately, these changes are particularly difficult to observe because they are very light and diffused.
Researchers believe that an AE could benefit from deep learning by collecting medical records from thousands of patients to create models and criteria for better identification of disease progression even at a very early stage. Currently, an AI model is based on brain imaging for a thousand patients and has managed to restore a correct diagnosis in old cases with a success rate of 100% (40 patients). The CA observed old images, dated 6 years before the disease (or not) of the patients in the group. A larger test scale could help develop the unit's reliability a little more.
This type of artificial intelligence could also be used within each brain imaging test to automatically examine each patient, its application does not require any particular interference from a human intruder.