Turkish scientists continue to make new discoveries one after another. One of the latest breakthroughs came recently and is related to lung cancer. In a significant breakthrough, they developed an AI tool to spot early-stage lung cancer by analysing speech!
Researchers at Ankara University trained the system to pick up subtle vocal changes—the kind most of us wouldn’t even notice—and it’s already showing 90% accuracy in early trials. The study, led by Dr. Haydar Ankışhan and presented by Dr. Yusuf Kahya, could pave the way for a noninvasive, fast, and low-cost screening method.
This groundbreaking approach comes from researchers at Ankara University and was recently presented by Dr. Yusuf Kahya at the Turkish Thoracic Society Congress in Antalya. The technology zeroes in on tiny vocal shifts — ones that most people wouldn’t notice — and it’s already showing about 90% accuracy in early tests.
The project was spearheaded by Dr. Haydar Ankışhan, with support from a team of medical professionals including Drs. Haluk Ulucanlar, İslam Aktürk, Kübra Alphan Kavak, Ulaş Bağcı, Bülent Yenigün, and Dr. Kahya himself. The researchers compared voice recordings from 50 people with early-stage lung cancer and 50 healthy individuals, feeding the data into an AI model trained to spot telltale patterns.
So, how does it work? Think of how your voice changes when you're sick — maybe a bit hoarse or weaker than usual. The AI is trained to detect those kinds of differences, except much more subtle and specific to lung cancer. What’s fascinating is that these vocal changes can occur well before any physical symptoms appear.
Dr. Kahya pointed out the potential impact of this method, especially since lung cancer is still the deadliest cancer worldwide. The biggest obstacle in fighting it? Diagnosing it early enough to treat effectively. That’s where this voice-based tool could really make a difference—it’s quick, noninvasive, and doesn’t rely on expensive or uncomfortable procedures.
However, the team quickly clarifies that this is still early-stage research. The initial results are promising, but before the system can be used in clinics, it must be tested on a much larger and more diverse population to confirm its reliability.
Still, the idea that something as simple as speaking into a microphone could help catch lung cancer early is a thrilling development in medical tech. If future studies back it up, voice analysis might one day become a regular part of health checkups — possibly saving countless lives in the process.
