0
0
NEW DELHI – A groundbreaking study suggests that subtle changes in a person’s voice, analyzed by artificial intelligence (AI), could become a non-invasive tool for the early detection of laryngeal cancer. The findings, published in the journal Frontiers in Digital Health, open a new frontier in diagnostic medicine, potentially offering a more accessible and rapid screening method than current invasive procedures.
Laryngeal, or voice box, cancer is a significant global health concern, with over a million cases and an estimated 100,000 deaths worldwide in 2021. Early diagnosis is crucial, as survival rates can be as high as 90% in the initial stages but drop significantly as the disease progresses. Current diagnostic methods, such as endoscopies and biopsies, are often uncomfortable, expensive, and not readily available in all healthcare settings.
The research, conducted by scientists at Oregon Health & Science University and Portland State University, analyzed over 12,500 voice recordings from 306 participants. The team used AI to examine various acoustic features, including tone, pitch, volume, and harmonic clarity. They discovered that specific acoustic patterns, particularly a change in the harmonic-to-noise ratio, were markedly different in men with laryngeal cancer and benign vocal fold lesions compared to those with healthy voices.
This “proof-of-principle” study indicates that AI could be trained to act as a vocal biomarker for cancer risk. While the current research showed clearer results for men, the researchers believe that with a larger, more diverse dataset, the technology could be refined to work effectively for all genders. The next steps involve clinical validation and training AI models on a wider range of voices to ensure accuracy and fairness across different populations.
The potential impact of this technology is immense. A simple voice recording could one day be used as part of a routine check-up, in telehealth appointments, or even through a mobile app to flag at-risk patients for further medical evaluation. This could lead to earlier diagnoses, improved treatment outcomes, and a significant reduction in the global burden of laryngeal cancer.
Disclaimer: This article is for informational purposes only and does not constitute medical advice. Consult with a healthcare professional for any health concerns or before making any decisions related to your health or treatment.
Reference: Phillip Jenkins, et al. “Voice as a Biomarker: Exploratory Analysis for Benign and Malignant Vocal Fold Lesions.” Frontiers in Digital Health. 2025.
The lifesaving device that can smell cancer
