Introducing TCR App Diagnostics
TCR App Diagnostics is a module within the mobile application built by TCR App Mobility. Our product is used for presymptomatic medical diagnostics of neurodegenerative diseases (PD, AD, ALS) on the basis of subtle driving anomalies.
We are turning the world’s road infrastructure into a neurological screening lab — identifying Parkinson’s disease approximately 5 years before the first tremor, using solely a smartphone. This process is made possible thanks to the non-invasive, ML-based analysis of nearly a hundred biometric data streams. Beyond neurology, our models are being adapted to detect other chronic conditions (e.g., glaucoma, diabetes, arthritis).
Technology
Current diagnostics of Parkinson’s disease rely on clinical motor symptoms (such as tremors) which typically emerge only after 60–80% of dopaminergic neurons in the brain have already irreversibly degenerated. At this stage, the window for neuroprotection is closed, leaving patients with purely symptomatic management. While the pharmaceutical industry is developing treatments capable of slowing Parkinson’s progression, these disease-modifying therapies (DMTs) are most effective only if administered before discernible symptoms appear. The economic burden of PD has surpassed €75 billion per year solely in the EU and the US. Broadening the scope, the aggregate impact of prevalent neurodegenerative conditions (PD, AD, ALS) is staggering, with nearly 75 million individuals affected and over €1.7 trillion in annual social costs.
Our algorithms are able to detect disease-specific changes in driving habits that are characteristic of chronic medical disorders (e.g., neurodegenerative diseases, as well as non-neurological conditions like diabetes, arthritis and glaucoma) 4–8 years before the onset of clinical symptoms. For example, in the case of Parkinson’s (approximately 5-year diagnostic lead time), we identify multi-modal indicators of prodromal neurophysiological decline and subclinical motor impairment (reflected in behavioural anomalies like erratic steering or delayed reactions) through analysis of the driver’s reward function evolution within a high-dimensional vector space (RKHS). By incorporating these unique patterns into our AIRL-based training workflow, we can uncover driving behaviour shifts associated with presymptomatic Parkinson’s disease.
Unlike traditional supervised learning, we employ adversarial inverse reinforcement learning (AIRL) to automatically calculate the individual reward function of a given person. The diagnostic power of our system lies in the ability to monitor the evolution of these reward functions over time. By analysing the longitudinal shifts of their distribution within a vector space (RKHS), we can identify the true onset of undesirable changes in human cognition. A statistically significant drifts in the RKHS potentially signals the initiation of a PD-related neurodegenerative process — detected at the stage (i.e., the first 5–10% of neuronal degeneration) when a patient is deemed healthy according to the most stringent medical protocols as of 2026.
The transformative breakthrough of this methodology lies in its departure from traditional, data-heavy diagnostic paradigms. Conventional AI approaches in PD/AD detection require massive, longitudinal datasets collected from already diagnosed patients — a process that is not only prohibitively expensive and time-consuming but also prone to significant geographic and ethnic biases. Our system bypasses these limitations by focusing on the fundamental allocation of cognitive resources. The user’s reward function is a mathematical description of how the human brain prioritizes sub-tasks to achieve an optimal driving policy. Because our model establishes an individual baseline in the RKHS, it eliminates the need for market-specific adjustment or large-scale data collection every time it is implemented in a different geographical region. This capability allows our technology to enter any global market and immediately define a local norm, ensuring objective diagnostic accuracy that is inherently immune to cultural or demographic variances in driving behaviour and ageing traits.
No action is required from the user beyond installing our application. It operates in the background (consuming minimal phone resources) and is fully compatible with older devices. It functions as a sophisticated data acquisition tool, passively collecting sensor data during the user's daily driving activities — whether the phone is mounted on the dashboard or kept in a pocket. Our application intelligently filters noise from relevant data streams.
Raw, anonymized data is securely transmitted to our cloud platform for regular processing (all complex ML operations are performed exclusively there). After a data collection period spanning several weeks to months, our system aggregates enough information to potentially generate a preliminary medical insight or diagnosis (if specific disease indicators are present). The results are then conveyed to the user and/or their healthcare provider. Our core technology (TCR App Mobility) was validated in an extensive market study, while its diagnostic algorithms are currently undergoing advanced optimization.
Market
The target market of TCR App Diagnostics consists of three different segments, starting with a high-value SOM focused on the R&D departments of approximately 100 global Big Pharma companies in Europe, United States and Japan. By providing accurate digital biomarkers for Parkinson’s detection approximately 5 years before the onset of clinical symptoms, we address the most critical bottleneck in drug development: the recruitment of presymptomatic patients for disease-modifying therapy trials. Our SAM (the next stage after SaMD/MDSW certification) encompasses the broader B2B healthcare sector, including hospitals, specialized medical centres and public health service providers across the EU, North America and Asia-Pacific markets.
Ultimately, our TAM represents the global healthcare ecosystem, the insurance industry and the population of individual drivers (B2C). We position our technology as a universal digital standard for the diagnosis and monitoring of chronic medical disorders, applicable to over 1.7 billion drivers worldwide.