The article AI detects throat cancer from the voice – pilot phase from 2028 appeared first in the online magazine BASIC thinking. With our newsletter UPDATE you can start the day well informed every morning.
Every year around 1.1 million people worldwide develop throat cancer. The diagnosis often occurs late and is very stressful for patients. A US research team has now developed an AI algorithm that is supposed to detect the disease based on voice alone.
In the future, the human voice could serve as a digital fingerprint for serious illnesses. An artificial intelligence should be able to detect throat cancer just by listening to speech. Loud current research results the voice hides subtle acoustic signatures that indicate early irregularities. The aim is to identify so-called vocal fold changes such as nodules or polyps in the bud.
Throat cancer is a massive global health problem, affecting approximately 1.1 million people annually. Around 100,000 people affected die every year as a result. According to experts, the main risk factors are smoking, high alcohol consumption and infections with human papilloma viruses (HPV). Traditional diagnosis via endoscopy or biopsies is considered invasive, painful and often difficult to access, which can delay vital treatment.
How does AI voice analysis for cancer detection work?
The Bridge2AI-Voice project investigates how computers can solve these complex biomedical problems. It is funded by the Bridge to Artificial Intelligence consortium of the US National Institute of Health (NIH). For the current study, a team led by Phillip Jenkins precisely analyzed a data set with 12,523 recordings from 306 participants.
Physical characteristics such as pitch, volume and general clarity of speech were measured. The researchers found that there were clear differences in the ratio of harmonic sounds to noise. Such patterns were particularly measurable in male patients with larynx cancer or benign vocal fold changes.
Artificial intelligence should be able to detect the smallest fluctuations in frequency (jitter) and amplitude (shimmer) that are barely noticeable to the human ear. This could enable a more objective assessment of individual cancer risk in routine clinical care.
When the AI diagnosis comes to the doctor’s office
No comparably clear trends have yet been identified among female study participants. However, the scientists assume that larger data sets will provide more reliable results in the future. A focus of the upcoming work will therefore be on the ethical procurement of multi-institutional data for broader training of the models.
Only after comprehensive validation in a clinical environment could the technology be used as a reliable medical tool. Jenkins estimates that corresponding AI tools for detecting changes in the vocal folds could enter the pilot phase in the next two years.
The method promises the advantage that it is completely contactless, cost-effective and can be used quickly in daily practice. Such digital biomarkers could bridge the critical waiting time for specialist appointments and significantly improve the chances of recovery through timely early detection. Depending on the stage and location of the tumor, the chances of survival with a timely diagnosis are currently between 35 and 78 percent.
Also interesting:
- Langdock: German alternative to ChatGPT and Co.?
- OpenAI vs. Anthropic: ChatGPT and Claude in direct comparison
- How to transfer your data from ChatGPT to Claude
- The LLM memory problem: Why AI often loses track
The post AI detects throat cancer from the voice – pilot phase from 2028 appeared first on BASIC thinking. Follow us too Google News and Flipboard or subscribe to our newsletter UPDATE.
As a Tech Industry expert, I am incredibly intrigued and impressed by the potential of AI to detect throat cancer from a person’s voice. This pilot phase in 2028 represents a significant step forward in the intersection of technology and healthcare, showcasing the power of AI in revolutionizing early cancer detection.
The ability of AI to analyze subtle variations in a person’s voice and accurately identify potential signs of throat cancer has the potential to save countless lives. Early detection is key in successfully treating cancer, and the use of AI in this capacity could lead to earlier diagnoses and improved outcomes for patients.
I am excited to see the results of this pilot phase and hopeful that it will pave the way for further advancements in AI-driven healthcare technology. The potential for AI to assist in early cancer detection is immense, and I look forward to seeing how this technology continues to evolve and make a positive impact on the field of healthcare.
Credits