Can Dogs Detect Cancer? Bengaluru Startup Combines AI And Canine Scent Detection With Promising Results

A Bengaluru startup is testing an unusual mix of biology and technology to spot cancer early. Dognosis is pairing trained dogs with artificial intelligence to study chemicals in human breath. Early trials suggest the method may provide a faster and cheaper screening tool. Doctors linked with the work say initial findings look promising for early detection.

The company reports that Phase-2 studies over two years covered nearly 1,500 people. Internal data suggests about 90 per cent accuracy in distinguishing cancer from non-cancer breath samples. The approach aims to support existing diagnostic methods rather than replace them. The team believes this could help more patients reach hospitals before the disease advances.

How dogs and AI to detect cancer work together in the lab

The screening process starts when volunteers breathe into specially designed face masks for a set time. These masks trap volatile organic compounds, or VOCs, that are present in exhaled air. Diseases such as cancer can change the pattern of these VOCs. The collected masks then move to the Dognosis laboratory for detailed assessment.

In the lab, trained dogs examine each breath sample inside a tightly controlled testing room. The setup includes multiple sensors and monitoring devices that track how the dogs respond. Their behaviour, such as pausing or signalling at certain masks, is recorded in real time. This information is then passed to AI systems for further analysis.

Training dogs and AI to detect cancer through smell

According to Akash Kulgod, CEO of Dognosis, the dogs learn to recognise specific VOC patterns linked to disease. "That mask contains volatile organic compounds or VOCs, which essentially carry signals indicating whether a person has a disease or not. The masks are brought to our lab, where a team of trained dogs evaluates them and identifies these compounds with high accuracy," he said.

Suba, who heads Research and Development at Dognosis, says the team focuses on strict testing standards. "Every sample presented here is presented in exactly the same way so that when the dog responds to it, we can record the outputs precisely. This data is then processed through AI systems trained using algorithms to convert what is usually considered subjective behaviour into objective scientific analysis," Suba said.

Clinical promise of dogs and AI to detect cancer

Oncologist Dr. Swaratika Majumdar has been associated with the work for about a year. "I have been working with diagnosis for the past year now, and it's a very interesting concept because it allows for an early, easy and inexpensive way of detecting cancers," Dr. Majumdar said. The doctor added that current studies indicate high sensitivity and specificity.

"There is a 90 per cent sensitivity and specificity, which means that if someone has cancer, the dogs are able to identify it around 90 per cent of the time," Dr. Majumdar added. Staff at Dognosis say this strengthens their belief that breath-based screening could support hospital-based cancer care.

Global interest in India’s dogs and AI to detect cancer

Srishti, an office associate at Dognosis, said the findings highlight both medical and local innovation value. "This is proof that cancer can be detected early just from breath. It is also proof that technology developed in India can contribute globally to early cancer detection," Srishti said, stressing the potential use in different health systems.

The company has invited trainers from abroad to prepare the dogs for this specialised work. "I came to India because we are doing something very unique here. I think this is among the most special detection works being done with dogs anywhere in the world, and it is possible here in India," said Edo, a trainer associated with the project.

India records several lakh new cancer cases each year, and many patients reach doctors at late stages. Researchers say tools that combine AI with biological detection methods such as trained dogs could support wider screening. If validated further, such systems may help hospitals offer quicker checks, especially where advanced lab facilities are limited.