AI may soon detect cancer by hearing patient’s voice

By Stephen Beech

A deadly form of cancer could soon be detected early by AI simply listening to a patient's voice, scientists say.

They believe state-of-the-art machine learning technology will be able to spot early voice box cancer from the sound of someone talking within a "couple" of years.

There were an estimated 1.1 million cases of cancer of the voice box, or larynx, worldwide in 2021, and around 100,000 people died from it.

Risk factors include smoking, alcohol abuse, and infection with human papillomavirus.

The prognosis for laryngeal cancer ranges from 35% to 78% survival over five years when treated, depending on the tumor’s stage and its location within the voice box.

Doctors say catching cancer early is "key" to a patient’s prospects.

Laryngeal cancers are currently diagnosed through video nasal endoscopy and biopsies, which doctors say are onerous, invasive procedures.

Now, researchers have shown that abnormalities of the vocal folds can be detected from the sound of the human voice.

Such "vocal fold lesions" can be benign, such as nodules or polyps, but may also represent the early stages of laryngeal cancer.

The American research team says their proof-of-principle results, published in the journal Frontiers in Digital Health, open the door for AI to recognize the early warning stages of laryngeal cancer from voice recordings.

Study corresponding author Dr. Phillip Jenkins, a postdoctoral fellow in clinical informatics at Oregon Health & Science University, said: “Here we show that with this dataset we could use vocal biomarkers to distinguish voices from patients with vocal fold lesions from those without such lesions.”

Dr. Jenkins and his colleagues are members of the ‘Bridge2AI-Voice’ project within the US National Institute of Health’s ‘Bridge to Artificial Intelligence’ (Bridge2AI) consortium, a nationwide push to apply AI to complex biomedical challenges.

In the new study, they analyzed variations in tone, pitch, volume, and clarity within the first version of the public Bridge2AI-Voice dataset, with 12,523 voice recordings of 306 participants from across North America.

A minority were from patients with known laryngeal cancer, benign vocal fold lesions, or two other conditions of the voice box: spasmodic dysphonia and unilateral vocal fold paralysis.

The research team focused on differences in several acoustic features of the voice, including pitch, jitter, variation in pitch within speech, shimmer, and the harmonic-to-noise ratio - a measure of the relation between harmonic and noise components of speech.

They found "marked" differences in the harmonic-to-noise ratio and fundamental frequency between men without any voice disorder, men with benign vocal fold lesions, and men with laryngeal cancer.

The team didn’t find any informative acoustic features among women, but they say it's possible that a larger dataset would reveal such differences.

They concluded that especially variation in the harmonic-to-noise ratio can be helpful to monitor the clinical evolution of vocal fold lesions, and to detect laryngeal cancer at an early stage, at least in men.

Dr. Jenkins said: “Our results suggest that ethically sourced, large, multi‑institutional datasets like Bridge2AI‑Voice could soon help make our voice a practical biomarker for cancer risk in clinical care."

Now that the proof-of-principle has been established, he says the next step is to use the algorithms on more data and test them in clinical settings on patient voices.

Dr. Jenkins added: “To move from this study to an AI tool that recognizes vocal fold lesions, we would train models using an even larger dataset of voice recordings, labelled by professionals.

"We then need to test the system to make sure it works equally well for women and men.

“Voice-based health tools are already being piloted.

"Building on our findings, I estimate that with larger datasets and clinical validation, similar tools to detect vocal fold lesions might enter pilot testing in the next couple of years."