When rescue dog Chloe is not tumbling across the grass on a farm outside Bengaluru, India’s technology hub, she is being trained to detect cancer through human breath with the help of artificial intelligence.
Equipped with a 3D-printed helmet and sensor-filled harness, Chloe is part of an experimental effort by Bengaluru-based startup Dognosis to harness dogs’ powerful sense of smell in developing a non-invasive cancer screening test.
A growing body of research suggests dogs can be trained to detect cancer and other diseases, including Parkinson’s disease and COVID-19.
Dognosis uses AI to analyse dogs’ brain activity, breathing patterns and body language, converting their responses to scents into data.
“In an hour, they can do thousands of sniffs,” Dognosis co-founder Akash Kulgod, 26, told AFP. “Each sniff can evaluate a sample.”
The company hopes to launch the technology commercially next year, which it says would make it the third company of its kind after firms in Germany and Israel.
Early detection can significantly improve cancer survival rates, but many patients in India are diagnosed only after the disease has progressed.
Dognosis believes dog-based screening could provide a simple and affordable first layer of testing that can be deployed widely to identify people who require further medical examination.
“It’s like a doggy ‘Iron Man’ suit,” Kulgod said.
Kulgod, who comes from a family of doctors and studied cognitive science at the University of California, Berkeley, said research indicates that diseases can alter chemicals released by the human body, producing distinctive odours.
Dogs have about 300 million scent receptors, compared with roughly five million in humans.
The testing process involves a patient breathing into a cotton mask for about 10 minutes. The mask is then sealed in a bag and transported to a laboratory, where trained dogs analyse the sample without coming into contact with the patient.
A recent Dognosis study involving more than 1,500 participants across six hospitals in Karnataka was published in the Journal of Clinical Oncology.
The company said the study recorded more than 90 per cent accuracy across seven major types of cancer, and it now plans to conduct tests in real-world programmes.
Cancer specialists in Bengaluru who were not involved in the project expressed cautious optimism.
“The study is promising, as an early signal,” said Santhosh Kumar Devadas, head of oncology at Ramaiah Memorial Hospital.
“But they need to do the study with a much larger group before we make any definitive conclusions,” he added.
Neelesh Reddy, a senior consultant oncologist at Manipal Hospitals, told AFP that “only time will tell how this technology will be useful to the actual patients”.
Dog ‘Work-Life Balance’
India has identified cancer as a major public health concern, with the country estimating that one in nine of its 1.4 billion people is likely to develop the disease.
Dognosis says its goal is to promote “proactive and preventative medicine” while improving communication between humans and dogs.
Most of its trained dogs are beagles, alongside some Malinois and Shepherd mixes. The animals are recruited through adoption or legal breeders.
Kulgod said only the best-performing dogs are selected, with each animal listed on the company’s website as an official “sniffer” alongside its photograph.
He added that the dogs have a “great work-life balance”, with their work limited to about 30 to 45 minutes a day.
At the company’s training centre, known as “Sniffspace”, dogs examine human samples on platforms fitted with automatic feeders that dispense treats as rewards.
Dognosis co-founder Itamar Bitan, 28, who spent four years in an Israeli special-forces canine unit, said the dogs are trained through positive reinforcement to identify a “specific funky smell” in exchange for a reward.
“Some of the dogs run directly to the feeder expecting rewards, some dogs freeze, some dogs will scratch the sample with excitement,” Bitan said.
“But, as long as the machine that learns how the dogs react to the sample is very consistent… we don’t really mind,” he added.
