For the millions of people navigating life after a stroke, the path to cognitive recovery is often a lonely one. Standard rehabilitation is typically anchored in a clinic, a scarce and expensive resource, leaving patients with limited, one-size-fits-all exercises to perform at home. A Polish startup called ABAStroke is trying to change that dynamic by bringing a deeply personalized, AI-driven therapist into the patient’s pocket. Its mobile application, which completed a randomized controlled trial last year, combines a behavioral science framework rarely used in stroke care with adaptive algorithms, aiming to offer a new standard for independent, at-home neurorehabilitation [ICT&health, May 2024].
The therapeutic wedge
ABAStroke’s core bet is on Applied Behavior Analysis (ABA), a methodology more commonly associated with autism spectrum disorder than with post-stroke cognitive deficits. The company’s algorithm uses ABA principles to break down complex cognitive tasks,like attention, memory, and executive function,into discrete, measurable steps. Machine learning then personalizes the difficulty and sequence of exercises in real time, creating what the company describes as “virtually unlimited exercise combinations” tailored to an individual’s progress [Perplexity Sonar Pro Brief, retrieved 2024]. This pairing of a structured behavioral framework with adaptive AI is its primary wedge against a field of more generic cognitive training apps. The company collaborated with digital health consultancy Revolve Healthcare to build the application, focusing on user experience and a pathway to medical device status [Revolve Healthcare, retrieved 2026].
Early clinical validation
The most compelling signal for ABAStroke’s approach comes from its clinical trial. The study, titled “Evaluation of the Effectiveness and Safety of ABAStroke,” measured cognitive function using the Montreal Cognitive Assessment (MoCA) [TrialX, 2023]. According to results reported by ICT&health, patients using the ABAStroke application showed an average improvement of 3.69 MoCA points, compared to a 2.00-point improvement in the control group receiving standard care alone. This translates to an 84% greater improvement attributed to the app [ICT&health, May 2024]. For a digital therapeutic seeking credibility with neurologists and payers, peer-reviewed data is the essential currency. This early validation suggests the underlying therapeutic model has merit, though larger, multi-center studies would be needed to cement its position in clinical guidelines.
The team and the Polish context
Operating from Kraków, ABAStroke is led by CEO Michał Ryś, a serial entrepreneur with a background in innovation management and prior experience at companies including DrOmnibus and Innovation Nest [RocketReach, retrieved 2026]. The Polish and broader Eastern European digital health scene has become a fertile ground for startups tackling specific, high-need clinical problems, often with a focus on rigorous science over rapid scaling. The company’s B2C model, targeting patients directly, presents both a challenge and an opportunity. It bypasses lengthy institutional sales cycles but must contend with patient affordability and discoverability in a crowded app marketplace. The lack of disclosed venture funding in the public record suggests the company may be bootstrapped or funded through grants, a common path for clinical-stage healthtech in the region.
The competitive landscape and risks
While ABAStroke positions itself as the first at-home digital rehab solution specifically for post-stroke cognitive deficits, it operates in a space with several inherent risks. The competitive landscape includes both general brain-training apps and a growing number of prescription digital therapeutics for neurological conditions. ABAStroke’s unique combination of ABA and AI is its differentiator, but it also requires educating both patients and clinicians on a novel methodology. Furthermore, the path to commercial scale with a B2C model in healthcare is notoriously difficult. Success likely depends on securing reimbursement from insurers or national health services, which in turn requires even more robust clinical evidence and regulatory approvals, such as a CE mark or FDA clearance.
- Clinical proof burden. The initial trial is promising, but adoption by rehabilitation centers hinges on replication in larger, published studies [TrialX, 2023].
- Commercial model friction. A direct-to-patient model faces hurdles in user acquisition and retention without the support of prescribing physicians or integrated care pathways.
- Category education. The application of ABA to stroke rehab is not mainstream, requiring clear communication to overcome clinical inertia.
The company’s next twelve months will likely focus on leveraging its trial data to seek partnerships with rehabilitation clinics or health systems, a move that could provide a more stable B2B2C channel. Additional funding rounds may also surface to support further clinical research and market expansion beyond Poland.
For patients living with the aftermath of an ischemic or hemorrhagic stroke, the standard of care for cognitive deficits remains fragmented. It often involves periodic, in-person neuropsychological therapy sessions supplemented by paper-based or simple digital exercises that lack personalization. The gap between intensive clinical intervention and the daily reality of recovery at home is vast and frequently filled with frustration. ABAStroke is attempting to build a bridge across that gap, one adaptive exercise at a time. Its success will not be measured in downloads alone, but in whether it can translate a novel algorithmic approach into measurably better, more accessible outcomes for a patient population that has long been underserved.
Sources
- [ICT&health, May 2024] First at-home digital rehabilitation of post-stroke cognitive deficits | https://www.icthealth.org/news/first-at-home-digital-rehabilitation-of-post-stroke-cognitive-deficits
- [Perplexity Sonar Pro Brief, retrieved 2024] ABAStroke company brief
- [Revolve Healthcare, retrieved 2026] Digital therapeutic for post-stroke rehabilitation | https://revolve.healthcare/projects/abastroke
- [TrialX, 2023] Evaluation of the Effectiveness and Safety of ABAStroke | https://www.trialx.com/clinical-trials/listings/263382/evaluation-of-the-effectiveness-and-safety-of-abastroke/
- [RocketReach, retrieved 2026] Michał Ryś Email & Phone Number | ABAStroke Co-founder and CEO Contact Information | https://rocketreach.co/michal-rys-email_110615029