A Farm Outside Bengaluru, Where Rescue Dogs Are Learning to Save Human Lives
On a farm on the outskirts of India’s tech capital, a rescue dog named Chloe splits her time between tumbling across the grass and something far less ordinary: learning to sniff human breath for early signs of cancer, with a little help from artificial intelligence.
Chloe is one of several dogs, including beagles, Labradors, and Dutch shepherds, at the centre of Dognosis, a two-year-old Bengaluru biotech startup betting that one of biology’s oldest tools, a dog’s nose, combined with modern AI, could reshape how cancer is screened for in a country where early diagnosis remains a persistent, often fatal challenge.
How the Testing Process Actually Works
The process is designed to be as simple as possible from a patient’s perspective, requiring no needles, scans, or invasive procedures of any kind.
- A patient breathes into a specially designed mask, called BreathEasy, for approximately 10 minutes.
- The mask captures volatile organic compounds (VOCs), microscopic chemical signatures released through breath that shift when disease processes, including cancer, are present in the body.
- Once sealed, the sample is transported, sometimes couriered, to Dognosis’s lab facility outside Bengaluru.
- Patients never interact directly with the dogs. The entire scent-analysis process happens separately, with trained dogs sniffing sealed, transported samples in a controlled lab environment.
The underlying science rests on a striking biological advantage: dogs possess around 300 million scent receptors, compared to roughly five million in humans, giving them a scent-detection capability that vastly outperforms even sophisticated laboratory equipment in certain contexts.
Turning a Sniff Into Data: The Role of AI
What separates Dognosis from a simple trained-animal detection method is the layer of sensor technology and machine learning built around the dogs themselves.
- Dogs are fitted with custom, 3D-printed helmets and harnesses packed with sensors, equipment the company has described in a way that draws comparisons to a canine “Iron Man” suit.
- These sensors capture brain activity, breathing changes, and body language as each dog processes a scent sample.
- Dognosis’s proprietary AI system, DogOS, analyses these subtle physiological and behavioural reactions, a pause, a shift in breathing, a change in posture, and converts them into objective, measurable, quantitative health scores.
- The company describes this combined approach as “olfaction AI“, merging canine intelligence, brain-computer interface technology, machine learning, and biomedical sensing into a single diagnostic-adjacent pipeline.
Co-founder Akash Kulgod described the sheer throughput this system enables: “In an hour, they can do thousands of sniffs. Each sniff can evaluate a sample,” he told AFP, underscoring just how much scale a single trained dog can offer once paired with the right sensor and AI infrastructure.
How the Dogs Are Trained and Cared For
For a publication focused on animal welfare, this is arguably the most important part of the story: how Dognosis treats the dogs at the centre of its research.
- Working hours are deliberately limited. Dogs work only 30 to 45 minutes a day, a structure specifically designed to keep them healthy, unstressed, and genuinely happy in their role.
- Training relies strictly on positive reinforcement. Co-founder Itamar Bitan, drawing on his background training detection animals in an elite military canine unit, explained that dogs are taught to detect what he described as a “specific funky smell,” and are rewarded when they successfully identify it.
- Rescue dogs are actively included in the programme. Chloe, a Labrador mix and a rescue dog herself, works alongside purpose-bred beagles and Dutch shepherds, meaning the programme isn’t reliant solely on specially bred detection animals.
This combination, short working hours, positive-reinforcement-only training, and the inclusion of rescued animals rather than exclusively bred ones, positions Dognosis’s approach as notably welfare-conscious compared to more intensive working-animal programmes seen in other detection-dog contexts globally.
The Scale and Scope of the Clinical Evidence So Far
Dognosis’s Phase 2 trial results represent the most substantial evidence base the company has published to date. The study involved more than 1,500 participants across six hospitals in Karnataka, with findings reported in the peer-reviewed Journal of Clinical Oncology, a detail that lends the research meaningfully more scientific credibility than an unpublished internal claim would carry.
The company reports an accuracy rate exceeding 90% across seven major cancer types, with its broader ambition extending to more than 20 cancer types in total. Company materials also indicate the detection approach has shown consistent results even in early-stage cancer cases, precisely the diagnostic window where earlier detection carries the greatest potential to improve patient outcomes.
Why This Matters Specifically for India
The core problem Dognosis is trying to address is a stark one: in India, more than 80% of cancer cases are diagnosed only after the disease has already advanced to a late stage, a statistic the company cites directly as central to its mission. Late-stage diagnosis dramatically reduces treatment effectiveness and survival outcomes, making any tool capable of shifting detection earlier a matter of significant public health consequence.
A genuinely non-invasive, low-cost, first-line screening tool, one that requires nothing more than a 10-minute breath sample rather than imaging, blood draws, or biopsies, could meaningfully expand access to early cancer screening, particularly in settings where the infrastructure and cost barriers of conventional diagnostic pathways remain significant obstacles for large segments of the population.
An Important Caveat: This Is Prescreening, Not Diagnosis
Despite the promising early results, it’s important to be precise about what Dognosis’s technology currently represents. As one detailed report on the company put it plainly: this is prescreening in clinical trials, not a proven diagnostic tool ready for standalone medical use.
Experts examining the field more broadly have stressed the need for larger, more extensive studies before drawing firm conclusions about real-world clinical reliability. Dognosis itself appears to recognise this distinction, positioning BreathEasy specifically as a first-line screening platform meant to flag individuals who may need further, conventional diagnostic follow-up, rather than as a replacement for existing cancer diagnostic pathways like imaging or biopsy.
What Comes Next
Dognosis is now pursuing further clinical validation and regulatory approval, with an eye toward potential commercial rollout. The company has stated it hopes to launch commercially in 2027, a milestone it says would make it only the third company of its kind in the world to bring this specific dog-and-AI cancer detection approach to market, following earlier entrants from Germany and Israel.
Beyond its initial published trial, the company has also been expanding its real-world testing footprint, with reports indicating breath-based detection trials extending to additional locations, including Belagavi in Karnataka, as it works to build a larger, more geographically diverse body of clinical evidence ahead of any regulatory submission.
Why This Story Resonates Beyond the Science
There’s something genuinely compelling about a story that combines cutting-edge AI, rigorous clinical trial methodology, and a rescue dog named Chloe, all working in service of catching cancer earlier in a country where late diagnosis remains devastatingly common. It’s a vivid illustration of how far the human-canine bond can extend when paired thoughtfully with modern technology, turning an ancient biological talent into a potential tool for saving human lives, without ever asking the dogs involved to sacrifice their wellbeing in the process.
As Dognosis continues toward its 2027 commercial launch target, its progress will likely be watched closely, both by the broader cancer diagnostics field and by anyone curious just how far a well-trained nose, and a well-designed AI system, can actually go.
Pets News Network is India’s first dedicated media platform for the pet and animal industry. For breaking global pet news, brand coverage, and advertising enquiries, contact: info@petsnewsnetwork.com
