A Russian neural network diagnoses dementia by voice: a new tool for early screening
Scientists from the Immanuel Kant Baltic Federal University have developed a multimodal neural network capable of detecting cognitive impairments through audio recordings of speech. This concerns the primary diagnosis of patients suspected of having dementia and other pathologies of the nervous system — at the earliest stages, when traditional methods are often powerless.
How Acoustic Analysis Works
Cognitive impairments at the initial stage can easily be confused with ordinary fatigue or age-related changes. However, voice parameters — intonation, pauses, speech tempo — change long before obvious clinical manifestations. The developed AI model simultaneously analyzes the semantic content of the response and 88 acoustic characteristics: voice tremor, volume fluctuations, and pause duration.
Unlike standard testing, where a doctor only evaluates the content of the response (e.g., describing a picture), the neural network follows the same scenario but supplements it with objective digital analysis. The system compares the recording with a database containing speech samples from healthy individuals and patients with confirmed diagnoses, and outputs the probability of pathology on a scale from 0 to 100%. Notably, specialized medical tests were not used during the neural network's training — the emphasis was placed specifically on acoustic features.
Limitations and Plans
It is important to emphasize: the development does not make final diagnoses. It is a primary screening tool designed to assist doctors in making clinical decisions. The team plans to turn the technology into a full-fledged medical product, for which it is already launching joint projects with major hospitals to collect a large-scale database of voice recordings. The research results were presented at the XV International Multiconference on Systems Biology in Novosibirsk.
Cryptalist Analysis: This is a vivid example of how AI is penetrating the most delicate areas of medicine. While the cryptocurrency market is in turmoil due to regulatory news, such breakthroughs in HealthTech show that fundamental artificial intelligence technologies continue to develop independently of market cycles. Investors should pay attention to startups implementing AI in diagnostics — this is one of the most sustainable trends of the coming decade.