Biomarqueurs vocaux : ce que notre voix nous dit sur notre santé

Veuillez noter qu’à l’exception du titre et du résumé, cet article est uniquement disponible en anglais.

• La santé digitale
• Santé de précision
• Unité de Recherche sur le Phénotypage Numérique Profond

Une équipe du LIH a publié une vue d’ensemble de l’utilisation du suivi de la voix dans le domaine de la santé numérique

Une voix peut révéler beaucoup de choses sur la santé d’une personne : est-elle forte ou au contraire, semble-t-elle faible ? Est-elle enrouée ? Existe-t-il des indications de douleur ou de fatigue ? Les technologies numériques modernes permettent depuis peu de détecter les moindres changements dans la voix. Cependant, des recherches supplémentaires sont à présent nécessaires pour rendre possible l’utilisation des résultats de ce suivi vocal à des fins médicales et de diagnostic. À cette fin, une équipe du Luxembourg Institute of Health (LIH) dirigée par le D^r Guy Fagherazzi, directeur du Département de la santé de la population et chef de l’Unité de recherche Phénotypage numérique profond a rédigé un article de synthèse sur le thème des « biomarqueurs vocaux ». Dans cet article, l’équipe de recherche décrit l’état de la technique de l’analyse de la voix à des fins de santé et de l’évaluation des enregistrements vocaux à l’aide de l’intelligence artificielle. Les scientifiques ont également décrit un pipeline dans lequel les techniques correspondantes peuvent être coordonnées et utilisées jusqu’à des applications médicales. Ils ont ainsi posé les bases essentielles pour l’évolution systématique de l’analyse vocale dans le domaine de la santé numérique et pour sa préparation en vue d’une utilisation dans la pratique clinique. L’article « Voice For Health:The Use Of Vocal Biomarkers From Research To Clinical Practice » a été publié le 16 avril dans la revue « Digital Biomarkers ».  

In the age of analogue medicine, a person would go to the doctor when he or she felt unwell. The doctor would perform an examination, make a diagnosis and prescribe a treatment for the patient. Until the next visit to the doctor, there was a period during which no one knew the patient’s exact state of health. But times are changing, as Dr. Guy Fagherazzi says: “We can now also use digital technologies to monitor a patient’s condition between two visits to the doctor – and intervene if his or her condition should deteriorate.” According to Fagherazzi, a key to this is the human voice. “If a person’s state of health changes, this immediately affects the voice,” the scientist says. The changes may be barely perceptible to the human ear. But digital technologies and artificial intelligence can measurably detect them as useful markers for diagnostic and medical purposes.

At LIH, this is an important new field of research. There are several projects addressing this topic, which hope to make digital voice analysis usable for combating COVID-19, among other things. “The first thing we did was to assess how far research in this field has already come,” says Fagherazzi. Together with his team and colleagues from the University of Luxembourg and the Luxembourg Institute of Science and Technology (LIST), he conducted a comprehensive literature review. The researchers learned which techniques are suitable for recording voices and how the data can be collected and stored. They compiled current methods for processing and evaluating voice recordings with the help of artificial intelligence, and identified which vocal biomarkers – which characteristics of the voice – can already be used to diagnose diseases and determine the state of health.

Describing the current health status is, however, only half the journey the LIH researchers decided to embark on. “We have also described in our paper how the different techniques need to be brought together and developed so that the use of vocal biomarkers becomes relevant for clinical practice,” Fagherazzi says.

We will now be taking these steps into practice within the framework of various clinical projects running at LIH and its cooperation partners

Two of these projects are related to COVID-19: in Predi-COVID, COVID-19 patients and their relatives are being systematically examined in order to identify biomarkers and risk factors associated with disease severity. In CDCVA, a project led by the University of Luxembourg and LIST in association with LIH, approaches are being researched to detect COVID-19 using cough and voice analyses. A third project, called CoLive Voice, will soon be launched to collect voice samples from volunteers all over the world. The goal of CoLive Voice is to advance voice-based diagnosis and symptom monitoring for a wide range of diseases, from cancer and diabetes to mental health and Parkinson’s disease.

Through all these projects, Fagherazzi hopes not only to gain new insights into digital voice monitoring, but also to ensure proximity to clinical practice:

Vocal biomarkers will only become useful if we have this connection with clinics.

For the future, he has three groups of key stakeholders in mind. The first is doctors who will be able to use voice analysis to monitor the condition and symptoms of their patients remotely, even when they are at home. The second is people who want to monitor their own current state of health using an app and voice samples. And the third group is pharmaceutical companies that can use the new techniques to capture better real-life data on the condition of their participants and on the tolerability and efficacy of new active substances in clinical trials. “We have now published a key paper,” Fagherazzi says. “But,” he adds with a wry smile, “more than anything else, the paper is the basis for a lot of work that now lies ahead of us »