D-rehabilitation of post-stroke cognitive deficits


Project info

D-rehabilitation of post-stroke cognitive deficits

ABAStroke is a mobile app for at-home neurological rehabilitation of post-stroke cognitive deficits. Combining Applied Behavior Analysis (ABA) methodology and machine learning (AI) algorithms, ABAStroke provides the patients with virtually unlimited exercise variations in an independent, personalized, and effective treatment. The exercise methodology created by ABA helps individuals return to normal functioning.


UX Design

UX Research

UI Design

Design Systems


Visual Identity

No items found.

Project summary

When does it start?

It all started in 2021 when the ABAStroke prototype was tested in a 4-week observational study with stroke survivors at the Department of Neurology in the University Hospital in Krakow, Poland. The study proved that ABA is an effective methodology for improving cognitive functions in patients, as confirmed by MOCA tests. Additionally, it demonstrated that individuals aged 60 and above can easily use mobile technologies. The effectiveness of the methodology and technology was further validated through user research, beta tests, and a clinical study. In 2022, we started working on the next iteration of the product.

For the user the process begins with an in-app assessment test, through which the app identifies the specific types and levels of cognitive deficits. Using an AI algorithm developed by ABAStroke, the app then generates a personalized rehabilitation path consisting of targeted exercises. As more data is collected from each session, the rehabilitation path is continuously optimized. By incorporating gamification techniques and following the ABA-derived method, the app motivates patients to use it regularly and provides them with the opportunity to improve at a faster rate.

We started the design collaboration by familiarizing ourselves with the problem and talking to specialists. The physical exercise list was very helpful in developing their digital counterparts. An adaptable design system that serves to present diverse resources, as well as a demanding target group, clearly set project goals - accessibility, clarity, larger clickable areas, and maximum resemblance to physical objects to facilitate adaptation among patients. Multiple tested product iterations have allowed us to refine not only the features, but also the interactions and communication within the application.

Exercise engines, rewards & statistics

The complete rehabilitation cycle encompassed several different exercise engines and an ever-expanding resource base. Personalized paths were created for each patient. Additionally, gathering a large amount of hospital patient data was crucial for the research. This data collection helped optimize AI algorithms to improve the effectiveness of treatment in the future.

The app is fully automated and designed to allow patients to work independently at home. It can predict the treatment already being provided to patients and make it more effective.

The rewards system and statistics were carefully crafted to provide strong incentives for patients to actively participate in their therapy sessions using the application. By incorporating gamification elements, such as virtual rewards and achievements, patients are motivated to consistently engage with the exercises and track their progress. Moreover, the system encourages patients to allocate a recommended duration of approximately 45 minutes each day for their therapeutic exercises, ensuring they dedicate sufficient time to their rehabilitation journey.

No items found.

We have been involved in the project for over a year. We would like to extend our special thanks to the several dozen professionals who have been engaged in the project, including the entire UX and Research team (Aga and Paulina), psychologists and clinical doctors (Tatiana, Ola, Kamila), the entire ML & AI team, developers, and of course, the founders (Michał, Dawid, and Basia).

Next case

You'll want to see this

Our clients

We’ve teamed up with some great people over the years