Laryngeal cancer, a significant health concern worldwide, may soon have an AI-powered tool to detect it earlier. Researchers from the US National Institutes of Health’s Bridge to Artificial Intelligence consortium have developed a method that analyzes voice recordings to distinguish between healthy individuals and those with laryngeal cancer.
The study analyzed 12,523 voice recordings from 306 participants, including patients with laryngeal cancer, benign vocal fold lesions, or spasmodic dysphonia. The researchers identified significant differences in the harmonic-to-noise ratio and pitch among men with healthy voices, benign lesions, and cancer.
These findings suggest that tracking changes in the harmonic-to-noise ratio could help monitor lesion progression and identify early-stage laryngeal cancer in men. However, more research is needed to confirm these results and develop a widely applicable AI tool.
The study’s authors believe that integrating voice-based health tools into triage systems could help identify patients at risk for laryngeal cancer based on their voice. While the current approach requires further validation, the potential for an AI-powered screening option is promising.
Source: https://www.technologynetworks.com/diagnostics/news/ai-analysis-of-voice-recordings-could-help-detect-laryngeal-cancer-403403