Can dogs help detect cancer? Study explores breath-based detection

Combining dogs' powerful sense of smell with advanced analytics, researchers are testing a non-invasive method to detect cancers early from breath samples

  From sniffing out explosives to tracking missing persons, dogs have long been trusted for their extraordinary sense of smell, and now that ability is being explored in the fight against cancer. Researchers are finding that trained canines can pick up subtle chemical changes in human breath that may signal the presence of disease. 

The study, titled “Canine Olfaction Combined With Bayesian Modeling for Multicancer Detection From Breath Samples: A Phase II Study in India” and published in the Journal of Clinical Oncology (JCO), examines how canine olfaction paired with technology could reshape cancer screening. Dognosis, the Indian startup behind the system, is developing an affordable, non-invasive breath-based multi-cancer detection test that combines trained dogs’ olfactory abilities with machine learning to generate measurable disease signatures.

 

What the study found

 

According to the study, trained dogs were able to detect multiple types of cancer from breath samples with over 90 per cent accuracy, including early-stage cases. This is significant because early detection remains one of the biggest challenges in cancer care. 

The study was conducted across six hospitals in India, with 1,502 people in the final test group, including 283 with biopsy-confirmed cancer and 1,219 healthy participants, as well as those with non-oncologic chronic diseases and benign biopsy results. 

The research highlights that:

• Detection rates exceeded 90 per cent in controlled settings
• Multiple cancers could be identified using a single breath sample
• Early-stage cancers, which are often missed by routine screening, were also detected

While these findings are encouraging, researchers stress that further real-world validation is essential before widespread clinical use. 

"The results are staggering yet not surprising to us - dogs have proved time and again that their ability to detect disease is superior to any other current options. We hope these results can be used to help advance and speed up diagnostic resources and screen many more people around the world," said co-author Claire Guest, Chief Executive and Director of Operations for the charity Medical Detection Dogs.  READ | Can microwave use lead to cancer? Here's what doctors say

How dogs detect cancer through breath

When cancer develops, it alters metabolism, and this leads to the release of volatile organic compounds (VOCs). "These VOCs create a unique "odor signature" or volatilome that trained dogs can identify, just like they can be trained to detect bombs and drugs. This signature is stable for a given disease across individuals," Akash Kulgod, Co-Founder and CEO, Dognosis told Business Standard. 

The process itself is simple and non-invasive, where a person breathes normally into a cotton face mask for about 10 minutes. The mask is then sealed and stored, and later evaluated by trained detection dogs in a controlled setting. Each sample is assessed independently by at least three dogs, and their responses are combined using an advanced Bayesian model that factors in each dog’s track record along with the participant’s background. 

Dogs are trained to recognise these scent patterns, much like they detect explosives or narcotics, and the method requires no blood tests, scans, or fasting.

Where AI fits into the picture

One of the key challenges with using animals in diagnostics is consistency, and this is where technology plays a crucial role. 

Kulgod explained that the team uses a Bayesian model to bring consistency to what would otherwise be variable biological responses, because it weighs each dog’s past performance along with patient-specific factors to standardise outcomes. 

He added that an AI-led training system, supported by a robotic platform, helps automate both training and evaluation, which in turn makes the process more scalable and suitable for handling large volumes of samples. 

With automation, even a small number of trained dogs could process large volumes of samples annually, making the model more practical than it might initially seem.  READ | Tumour vs cancer: What's the difference and why these terms confuse many

Potential benefits over traditional screening

Current cancer screening methods can be invasive, expensive, or limited to specific cancer types, while breath-based detection offers a simpler alternative. Some potential advantages include:

• Non-invasive and painless testing
• Faster results compared to lab-based diagnostics
• Ability to screen for multiple cancers at once
• Lower infrastructure requirements

However, experts caution that this method is not yet a replacement for established diagnostic tools, but it could become a useful first-line screening option.

What happens next

The research is now moving beyond controlled environments into real-world pilot studies, where scientists aim to validate results across larger and more diverse populations. This step is important because factors like diet, environment, and other health conditions can influence breath composition. 

Kulgod noted that the team is already testing the system beyond the lab, saying, “We are running real-world pilots to generate prospective validation of our system.” He added that future plans depend on evidence, stating, “Once we have sufficient evidence, we will be aiming to commercialise next year.” 

While more evidence is needed before clinical adoption, this approach signals a shift towards simpler, more accessible screening tools. 

For more health updates, follow #HealthwithBS

This report is for informational purposes only and is not a substitute for professional medical advice.