Abstract
Currently, there are virtually no objective diagnostic signs (markers) to diagnose neurodegenerative diseases, that is why it may take months of constant monitoring of symptoms to establish a reliable diagnosis. Now there is an increasing number of reports that artificial intelligence can provide a more accurate diagnosis of neurodegenerative diseases by identifying specific diagnostic features from electroencephalography data, neuroimaging, and wearable devices and smartphones. This study reviews the application of artificial intelligence to the diagnosis of neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease, and Huntington's disease. Using the international databases Scopus and PubMed, as well as the Russian Science Citation Index, we analysed studies that used artificial intelligence methods to detect, monitor, or control the progression of neurodegenerative diseases. The primary focus is on analysing electroencephalography data, neuroimaging data, and data from wearable devices and smartphones. The use of the latter may allow screening, diagnosis, and monitoring of the disease at home with a minimum economic cost. Artificial intelligence can be a useful tool for early, accurate, and non-invasive diagnosis of neurodegenerative diseases, as well as for assessing the effectiveness of treatment and predicting the course of the disease. However, for the widespread implementation of artificial intelligence in clinical practice, several issues related to the quality, availability, and standardization of data, validation and interpretation of results, ethical and legal aspects need to be resolved. The use of artificial intelligence requires both specialists from the IT industry with a deep understanding of the types and kinds of neural networks and physicians with fundamental knowledge of neurology, psychiatry, molecular biology, and biophysics.
Publisher
PANORAMA Publishing House