Developing AI cognitive health monitoring software through computer usage habits

Mantis Biotech, a U.S. tech company, has launched Parva, an artificial intelligence (AI) software that monitors users' cognitive health by tracking keyboard typing patterns and mouse movements on computers in everyday life. Instead of relying on periodic medical assessments, Parva builds continuous cognitive profiles for individuals, documenting attention trends, daily rhythms, and work patterns to detect subtle changes in cognitive function.

The core idea is to compare users to their own historical patterns over time rather than benchmarking against population norms that are gender-biased and fail to accurately represent individuals.

Overcoming limitations of traditional cognitive health monitoring methods

Current assessment methods primarily rely on regular medical appointments, usually occurring once a month with inconsistent intervals. This approach makes it difficult to track changes in cognitive status, as it only reflects the condition at a specific point in time, lacking continuity.

For women, this is even more disadvantageous because symptoms often fluctuate with monthly hormonal cycles. For example, emotional self-regulation varies with sex hormones, making it challenging to accurately diagnose conditions such as perimenopausal disorders, Alzheimer's disease, or long COVID—illnesses that significantly affect women more frequently.