According to a study published in the journal Radiology, artificial intelligence can help predict Alzheimer’s disease, a disease where early diagnosis can be pivotal for the introduction of treatments and interventions.
Diagnosing Alzheimer’s is no easy task and, so far, research has only managed to link the disease process to metabolism changes shown by glucose uptake in certain regions of the brain but even knowing this, the changes are, more often than not, difficult to recognize.
Jae Ho Sohn, M.D., from the Radiology & Biomedical Imaging Department at the University of California in San Francisco (UCSF) said about these specific markers that the “differences in the pattern of glucose uptake in the brain are very subtle and diffuse. People are good at finding specific biomarkers of disease, but metabolic changes represent a more global and subtle process.”
To make the process easier, researchers decided to allow artificial intelligence in the game – they trained a deep learning algorithm on an imaging technology called 18-F-fluorodeoxyglucose positron emission tomography (FDG-PET). During this type of scan, a radioactive glucose compound is injected into the patient’s blood. The PET scans can then measure the FDG uptake in the brain cells, which is an indicator of metabolic activity.
In addition, the researchers accessed the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset and fed it to the deep learning AI. This way, the AI managed to see all the patterns that corresponded to Alzheimer’s disease.
After the deep learning process was completed, the researchers have tested the algorithm on a set of 40 imaging exams from 40 patients the AI never studied before. The AI achieved 100% accuracy at detecting the disease more or less six year prior to the diagnosis.
At the moment, the researchers are working on improving the quality of the AI even more and, in the future, thanks to this software, perhaps we could see an end to a disease that affects millions all across the world.