This advancement, developed in Alicante, utilizes advanced natural language processing (NLP) and deep learning technologies to identify patterns of cognitive decline through acoustic and linguistic signals present in speech.
The tool, named Artificial Intelligence Platform for Early Detection of Alzheimer's Disease through Voice (IAEAV), aims to facilitate early diagnosis, which is key to improving the disease's progression.
“"Current treatments are more effective when applied in initial stages, so detecting Alzheimer's in time can delay its progression and improve the quality of life for both patients and their caregivers."
The system is supported by studies showing that the first neurological changes can be reflected in language, with signs such as prolonged pauses, reduced syntactic complexity, or errors in verbal fluency. To collect this data, a simple mobile application has been designed, allowing users to record their voice in different contexts, such as reading texts, spontaneous narratives, or answering questions.
Recordings are subsequently analyzed to extract acoustic features like tone or intensity, as well as linguistic aspects related to semantic richness. This data is evaluated using deep learning models trained with representative databases.
One of the most relevant aspects of this project is its accessibility, as the application is designed for use in both clinical settings and at home, reducing barriers and facilitating access for people with fewer resources. The project has been funded by the Generalitat Valenciana through Next Generation funds.
