Meet Blat: The Barcelona dog that can detect lung cancer from sniffing a person’s breath

Fiona Govan
Fiona Govan - [email protected]
Meet Blat: The Barcelona dog that can detect lung cancer from sniffing a person’s breath
Blat and his owner (and trainer) Ingrid Ramón. Photo: Ingrid Ramón

Every dog owner will tell you that their four-legged friend is brimming with special qualities. But in the case of this labrador-cross, it wouldn’t be empty boast.


For Blat is a pioneering dog with a powerful nose. He is one of just a handful of canines in the world – and the only one in Spain – that has been trained to detect lung cancer.

And he can do so from just a whiff of the breath of someone suspected of having the disease.

According to a study published in the European Journal of Cardio-Thoracic Surgery, Blat had a success rate of 95 percent in detecting cases of lung cancer and was even able to do so in very early stages – identifying cancer when the tumour was a mere four millimeters in diameter.

“The dog seems able to pick up everything, even when the tumours are very small, it’s astonishing,” wrote Angela Guirao, co-author of the study which was carried out at Barcelona's Hospital Clinic.

“It’s basicially a game for Blat,” his owner and trainer Ingrid Ramón, who adopted him at eleven months when he needed rehoming, told The Local.

“I recognized him as a labrador retreiver mixed with an American Staffordshire pit bull and that is a cross that makes for easy training. He was a little hooligan because he hadn’t been trained properly but I saw that he had great potential and took him on.”

After months of training, Blat, now three-a-half years old, can sniff breath samples - which have been hermetically sealed and are then opened and placed within a wooden box with a hole to amplify the scent.

Within seconds the dog can detect lung cancer and indicate it by sitting next to the sample.

Watch Blat identify cancer samples:

“He’s been trained to pick up the sample through positive reinforcement for which he gets a treat.”

”Blat’s results indicate that there are molecules that are specific to lung cancers and that some of these molecules are detectable in the exhaled air," explained Laureano Molins, a thoracic surgeon at Hospital Clinic and co-author of trial.

"These are spectacular results," said Molins explaining the need to improve early diagnosis as "75 percent of lung cancer cases are diagnosed when the disease is already advanced."

However, the study's ultimate goal is not to introduce teams of diagnostic dogs in oncology but rather to aid development of a test that can identify the molecules specific to certain types of cancer.

"At the moment the olfactory skills of a dog are superior to any technology we have today," said Molins. "Our goal now is to identify the molecules (detected by Blat) and develop a diagnostic test that acts as an electronic nose."

But although Blat is a pioneer, he is in good company as his owner runs Argus Detection Dogs, which trains dogs to specifically indentify types of illness.

Among those trained by Ramón are dogs which can detect hypoglycemia - when blood sugar drops to a dangerous level - in Diabetes sufferers, as well as dogs that can warn of an imminent attack in those with epilepsy.

She also has trained dogs who accompany children with autism or reduced mobility.

"Blat is incredibly special. He is a real sweetheart and loves people" boasts Ramón, but she admits that's not just on account of his skilfull nostrils.  

"Honestly, most dogs have the same olfactory capability and can be trained like this. What makes Blat exceptional is  his willingness to work. He is a mix of Labrador and American pitbull and that means he loves to eat, he loves to play and he has incredible energy."


Join the conversation in our comments section below. Share your own views and experience and if you have a question or suggestion for our journalists then email us at [email protected].
Please keep comments civil, constructive and on topic – and make sure to read our terms of use before getting involved.

Please log in to leave a comment.

See Also