Phone Becomes a Caregiver: Russian Students Create an App for People With Dementia
Russian students have created an app for people with dementia and their families. It analyzes geolocation and accelerometer data, learns to distinguish normal behavior from potentially concerning patterns, and instantly alerts relatives with location coordinates if an incident occurs. The project has already been included among Russia’s leading scientific achievements.
People living with Alzheimer’s disease and other neurodegenerative disorders can gradually lose the ability to care for themselves independently. They may leave home to go to a store and fail to find their way back. They may set out to visit familiar friends and end up in a different neighborhood. They may forget they took medication an hour earlier and take another dose. Around 1.5 million people in Russia are currently living with Alzheimer’s disease. Behind every one of those patients stands a family coping with constant stress and concern for a loved one.
How can families make life safer for someone they care about? That question led Aleksey Osipov and Yevgeny Shirochkin, students at the Pusk (Launch) Center of the Moscow Institute of Physics and Technology (MIPT), to develop a solution. Drawing on personal experience with people affected by neurodegenerative diseases, they set out to create a reliable digital assistant for caregivers. The result was Memory Care, a mobile application designed to monitor patients, provide an added layer of protection, and warn families about potential dangers.
How Does the App Work?
Memory Care consists of two components. One is installed on the patient’s phone, while the second runs on the caregiver’s device. A self-learning model operates in the background on the patient’s smartphone. It analyzes two types of information: geolocation data, which shows where the person is, and accelerometer readings, which reveal how the phone is positioned, whether it is moving, and whether it may have fallen.
The algorithm learns to recognize normal behavior patterns. It identifies routine activities such as walking to a local store, taking a stroll around the neighborhood, or visiting a clinic. Once those patterns are established, it begins detecting anomalies: wandering in circles, spending extended periods in unfamiliar locations, or making sudden and illogical route changes. It can also detect periods of inactivity if the phone remains in the same position for an unusually long time despite conditions suggesting the person should have returned home. Most importantly, it recognizes falls by identifying a sudden change in device orientation followed by immobility. When an incident is detected, relatives immediately receive a push notification containing the patient’s coordinates.
Adapting to the Individual
Memory Care’s key advantage is its ability to learn. During the first one to two weeks, the model studies the habits of a specific user. It tracks where the person usually goes, at what times, and which routes are typical. After that, it can distinguish ordinary behavior from potentially dangerous situations with increasing accuracy. The system continues learning over time, improving both precision and reliability with every passing day.
Its functionality can be compared to a GPS tracker, but Memory Care goes considerably further. Rather than simply showing a location, it acts as an analytical system capable of recognizing that something may be wrong before a family member has reason to become alarmed.
What Comes Next?
The application currently tracks wandering behavior and falls. Over the next six months, the developers plan to add detection of freezing episodes, a feature that could be particularly important for people living with Parkinson’s disease. Future versions are also expected to include medication-management tools, providing reminders when it is time to take medicine and recording whether doses have already been taken.
The team plans to launch subscription sales through partner clinics and insurance companies. Longer term, the roadmap includes releasing the app through the App Store and Google Play for general users, as well as developing versions for smart bands and smartwatches.
Peace of Mind Instead of Constant Anxiety
For families caring for elderly relatives, the application could become an indispensable support tool. It would allow caregivers to focus on work and daily responsibilities with greater confidence, knowing that a phone can issue an alert if something goes wrong. For patients, the app offers a degree of independence without the uncomfortable feeling of being watched by a warden. All that is required is carrying a smartphone with the application installed. The technology could also significantly ease the burden on volunteer search-and-rescue organizations. Fewer people with cognitive impairments may go missing without anyone knowing where to begin looking.
A Universal Solution
With Memory Care, MIPT students have addressed a real-world challenge affecting millions of families. The project has already been recognized among Russia’s most significant scientific achievements. The estimated initial market size in Russia is up to 21 million rubles per month (approximately $270,000) and about 42,000 users, with growth potential reaching 200,000 people. The project already has a development roadmap and a financial model, giving it a realistic path toward mass adoption.
Dementia is, in fact, a global challenge. Tens of millions of people worldwide live with Alzheimer’s disease, and their families face the same difficulties: patients become lost, wander away, and encounter dangerous situations. The disease spares neither ordinary people nor public figures. Sean Connery, Margaret Thatcher, and Bruce Willis all experienced forms of dementia or Alzheimer’s disease. That gives the project substantial social importance and long-term potential. Memory Care could evolve into a universal tool for supporting elderly care around the world.
