In the landscape of technological innovations applied to healthcare, the role of artificial intelligence (AI) is becoming increasingly important, especially in its ability to collect and interpret complex biometric signals. We discuss this with Andrea Buccoliero, Vision & Innovation Manager of the R&D Department at GPI, who introduces us to a cutting-edge project, Talìa, capable of detecting disease biomarkers through voice analysis.
Dr. Buccoliero, can you explain how the voice can become a diagnostic tool?
Yes, the voice is an extremely rich source of information. It is a systemic characteristic of human beings, influenced by a multitude of physical and psychological factors. Thanks to AI algorithms, we can now analyze it in great detail, identifying biomarkers that signal the presence of specific diseases, such as cardiovascular, neurodegenerative diseases, or even burnout. The Talìa project, which we are working on, aims to develop and validate algorithms that can detect these signals in real-time.
What is the main ambition of the Talìa project?
The main goal is to provide an advanced tool that, using AI, can not only detect signs of disease but also become a fundamental resource for doctors during patient follow-up. In other words, by recording vocal patterns associated with a specific disease, the doctor can monitor the progression of the disease over time.
How does this technology fit into the context of telemedicine?
The advantage is that all of this can be done remotely. The use of voice allows for more accurate staging of the disease and enhanced follow-up, without the need for the patient to physically go to the hospital. This becomes crucial in telemedicine or teleassistance settings, where the doctor, thanks to the collected data, can manage remote monitoring more effectively and personalize care based on the progression of the disease.
Another area where TALIA promises to be innovative is in burnout prevention. How does it work in this case?
Exactly, another goal of Talìa is to use Speech Emotion Recognition technologies to detect signs of chronic stress, what we know as burnout. The idea is that small repeated stress events can be detected in the tone of voice, even before they become evident through physical symptoms. Chronic stress is known to be linked to a range of both physiological and psychological illnesses, such as cardiovascular disorders or depression. Detecting these signals early could open new avenues in burnout prevention, also improving corporate wellness.
So, could TALIA become a prevention tool for workers as well?
Absolutely. We are collaborating with the University of Verona, with professors Andrea Ceschi and Riccardo Satori, to understand how early stress detection can help improve employee well-being. If we can identify signs of prolonged stress in the tone of voice, we can intervene before more serious problems arise. In this sense, Talìa has the potential to be useful not only in the medical context but also in the corporate environment to monitor and prevent burnout.
Gpi4AI_Diagnosys: new frontiers for diagnosis and monitoring
Gpi4AI is the new offering line that encompasses all the AI Agents our group is working on. In particular, Gpi4AI_Diagnosys represents a significant step forward towards increasingly personalized and proactive medicine, capable of leveraging advanced technologies such as AI to improve disease diagnosis and monitoring. Whether it’s helping doctors diagnose diseases or preventing burnout in corporate settings, voice analysis opens new possibilities for enhanced telemedicine, combining innovation and patient care in a more effective and sustainable way.
