Bengaluru: While we all know that dogs are sensitive to smell, how they can be trained to detect cancer is an interesting research that Bengaluru-based startup Dognosis has taken up.
According to researchers, cancer alters cellular metabolic processes, leading to changes in the body’s volatilome - the complete set of volatile organic compounds (VOCs) produced by an organism.
“These alterations create unique ‘odour-prints’ that have been detected in biological samples such as breath, urine or sweat. Breath samples are particularly promising for cancer detection due to their non-invasive nature and the rich variety of almost 1,500 VOCs that they contain,” Dr Minal Dakhave, medical affairs lead, told DH.
Considering that the dogs possess an olfactory system which is way more sensitive than human noses or current artificial sensors and the fact that they can be trained to communicate their analysis has made dogs the ideal candidates for detecting subtle changes in VOC profiles associated with cancer.
As part of its research, the team collected VOCs from patients through a cotton mask. This would mean that the patient will have to only breathe through a mask, avoiding the invasive procedures, once research is approved and scaled up for real time use.
“During the research phase, participants were instructed to breathe normally through the mask for 10 minutes, taking a deep breath every minute to ensure adequate sampling of air from the alveolar region,” noted Sahana Shanbhag, chief clinical coordinator.
These masks were stored safely and passed on to the dogs for identification.
Based on the behaviour of the dogs, the handlers would call out the analysis.
The dog’s behavioral responses were recorded via motion sensors, video data, and electroencephalography (EEG) systems.
To ensure scientific robustness, the study was designed as a double-blind trial— neither the dog handlers nor the researchers analysing the results knew the health status of the participants.
Validity of findings
This approach minimises bias and strengthens the validity of findings, offering rare empirical insights into canine olfaction and cancer-specific VOC signatures.
The study also explores how factors like diet, environment and comorbidities might influence individual VOC profiles, helping differentiate cancer-specific patterns from background biological noise.