Artificial intelligence, which can detect the emergence of early Alzheimer's disease, was developed by researchers at the Laboratory of Computer Science and Artificial Intelligence of the Massachusetts Institute of Technology, USA. Today, there are no precise methods for diagnosing Alzheimer's disease. There is no single test, and brain scanning does not always help to determine if the patient has this disease. At present, doctors evaluate several factors, including the medical history of the patient and the observations made by family members or medical specialists. And this despite the fact that disorders in the brain can cause subtle changes in the behavior and habits of sleep years before the patient begins to feel the loss of memory.
According to researchers at the Laboratory of Computer Science and Artificial Intelligence of the Massachusetts Institute of Technology, an artificial intelligence system can recognize such changes at an early stage and identify patients who have a severe form of the disease. They have developed an AI-based device that is able to trace the patient's behavior imperceptibly and analyze it for a long time.
This device is a flat rectangular box with rounded edges that hangs on the wall of the room where the patient lives and does not show any activity. However, this device knows when a patient gets out of bed, dresses, goes to the window or goes to the bathroom. It may notice that a person is sleeping or registering a fall. All this is done with the help of low-power wireless signals that record the speed of movement of a person, his gait, sleep patterns, location and even how he is breathing. All this information is automatically loaded into the cloud system, where it is analyzed by the algorithm based on machine learning, taking into account the already accumulated data about this patient.
Signals of this device, which are a thousand times less powerful than WiFi, allow you to record all changes within a radius of 10 meters, including the bodies of people. Any movements, even the weakest ones, like breathing, cause changes in the reflected signal and are fixed. The algorithm of machine learning analyzes all these data and is able to distinguish between them. The algorithm allows you to notice deviations from the usual patterns of human behavior that may indicate psychomotor agitation, depression and sleep disorders. It can also mark if a person begins to repeat certain behaviors several times a day. All these are classic symptoms of Alzheimer's disease.
Dina Katabi and her team at MIT’s Computer Science and Artificial Intelligence Laboratory initially developed the device as a fall detector for older people. But they soon realized it had far more uses. If it could pick up on a fall, they thought, it must also be able to recognize other movements, like pacing and wandering, which can be signs of Alzheimer’s. Use such a device can be at home in elderly people, and in nursing homes in order to monitor the behavior of people at risk. For people who already suffer from Alzheimer's disease, this technology can help to adjust the treatment, taking into account changes in a person's life.