The BBHI joins the ADDF’s SpeechDx, an international initiative for the study of voice as an early biomarker of Alzheimer’s

The BBHI has joined SpeechDx, a new study launched by the Alzheimer’s Drug Discovery Foundation (ADDF)’s Diagnostics Accelerator (DxA), which aims to enable the development of voice and speech biomarkers for Alzheimer's disease and other neurodegenerative pathologies.

The project will collect data on the voice and cognitive status of 500 BBHI volunteers aged between 45 and 70 for three years. Participants will receive a tablet through which they will record their voice every three months, performing tasks such as describing an image or repeating a story. These voice recordings will be paired with data on the participants' health that has been collected since the start of the BBHI. This paired voice and health data will be cleaned of all personal data and harmonized with the data from the rest of the SpeechDx cohorts, so that all SpeechDx data can be analyzed together.

The Institut Guttmann will contribute healthy subjects to the study. Data provided will be used to characterize the normality values in comparison with those of the other cohorts, provided by the Barcelonaβeta Brain Research Center (BBRC) and the ACE Alzheimer Center (Barcelona, Spain), Boston University (Boston, USA), and Emory University (Atlanta, USA), among others.

Main threat to public health

According to the World Health Organization, neurological and mental disorders are currently the main threat to public health: pathologies that affect the brain are already more prevalent than cancer and cardiovascular diseases. This is largely due to the aging of the population since advanced age is the main risk factor for the development of these diseases. The United Nations predicts that by 2050 there will be twice as many people in the world over the age of 65 as under the age of 15.

In this context, early detection of neurodegenerative pathologies is key to the development of intervention strategies that minimize their impact. In recent years, analysis of voice and speech has emerged as a possible biomarker of the preclinical stages of Alzheimer's disease.

“Thanks to new technologies, researchers can collect and analyze information about speech and language remotely and efficiently over time,” explains Javier Solana, Research Coordinator at the Institut Guttmann. Solana is the Co-Principal Investigator of the project along with Postdoctoral Researcher Gabriele Cattaneo, PhD. “These tools can measure acoustic, lexical, and syntactic aspects of speech, as well as written language characteristics associated with Alzheimer's disease. We can glean insights from not just contextual or semantic information, but also aspects such as tone, pauses, tremor or amplitude,” he continues.

However, for the analysis of these factors it is necessary to develop artificial intelligence and machine learning-based algorithms that analyze voice and speech with great precision. To this end, SpeechDx will create a repository of samples representing different accents, languages, speech, and language components, among participants with different stages of the brain health spectrum—from no Alzheimer’s disease, to mild cognitive impairment and finally early Alzheimer’s.

“Speech is a complex cognitive process that contains important information about how your brain is functioning and scientific evidence shows us speech may hold the key to early, accurate, and non-invasive detection of Alzheimer’s disease,” says Howard Fillit, MD, Co-Founder and Chief Science Officer of the ADDF. “This study will help develop and validate voice-based biomarkers, expanding our existing arsenal of neuroimaging, peripheral blood, and digital biomarkers—all of which are crucial to delivering the right drugs to the right patients at the right time.”