Bengaluru Startup Combines Dogs and AI to Sniff Out Early Cancer

A two-year-old Bengaluru startup, Dognosis, is pioneering an innovative approach to early cancer detection by merging the superior olfactory capabilities of trained dogs with advanced artificial intelligence. The company aims to provide a non-invasive, cost-effective prescreening tool that could significantly improve cancer detection rates, particularly in regions with limited access to traditional diagnostic methods.

Dognosis's method is deceptively simple for patients. Individuals breathe into a mask, which is then securely transported to the startup's facility, a farm nestled outside Bengaluru. Here, a specialized team of dogs, including beagles, labradors, and Dutch shepherds, are trained to sniff these breath samples for subtle chemical signatures indicating the presence of tumors.

As the dogs interact with the samples, an intricate system of sensors and AI monitors their physical and neurological responses. This includes tracking their movements, respiration patterns, brain activity, and body language. A simple pause or a tail wag from a dog is then translated by the AI into structured data, allowing for a standardized and repeatable analysis. This entirely sidesteps the need for invasive procedures like biopsies, blood draws, or complex imaging scans often associated with cancer detection.

Founded by Akash Kulgod, Dognosis states its technology can target more than 20 types of cancer across seven categories, including prevalent forms like breast, oral cavity, lung, and cervical cancers. The company reported impressive preliminary results from its Phase 2 trials, published in the Journal of Clinical Oncology, showing approximately 90% sensitivity for early-stage cancers. Additionally, Dognosis claims that dogs trained on 10 specific cancer types were able to identify an 11th type they had not previously encountered. It is important to note that these accuracy and detection figures are currently reported by the company and its researchers, awaiting independent confirmation.

Addressing India's Cancer Challenge

The Dognosis approach is particularly relevant given India's significant cancer burden and low screening rates. In 2024, the country recorded about 1.5 million new cancer cases and over 900,000 deaths, making it one of the heaviest caseloads globally. Despite this, screening rates remain remarkably low, with roughly 1.9% of women screened for cervical cancer and less than 1% for breast or oral cancers. This gap is largely attributed to factors such as high costs, geographical distance to facilities, and patient discomfort with invasive procedures.

A cheap, non-invasive prescreening tool could dramatically expand access to early detection for a large population that the traditional healthcare system currently struggles to reach. This is precisely the critical market Dognosis aims to serve with its innovative solution.

Funding and Future Outlook

Dognosis secured a $1.5 million pre-seed funding round in 2024 and has since closed another undisclosed round, attracting notable investors like Accel India and Caffeinated Capital. The startup has an ambitious roadmap, planning to launch an at-home, multi-cancer prescreening test in India next year, with an estimated price point of a few thousand rupees.

Currently, a Phase 3 trial is underway, initiated in April, with plans to enroll approximately 10,000 participants across 10 hospitals. The company targets performing a million tests annually with a one-week turnaround, supported by a team of around 30 highly trained dogs. Looking further ahead, Dognosis has penciled in the establishment of a US facility for late 2027 or early 2028.

Important Caveats and Regulatory Landscape

While promising, there are important distinctions to highlight. Dognosis operates as a prescreening tool, designed to flag individuals who should seek further medical evaluation, rather than providing a definitive diagnosis. This distinction is crucial and will require careful communication from the company to patients.

Kulgod noted a regulatory nuance in India: because Dognosis offers a prescreening tool and not a formal diagnostic device, it does not require Indian regulatory approval to launch. While this lowers the barrier to market entry, it also implies that the technology may reach patients without undergoing the rigorous approval processes typically required for official diagnostic medical devices. Furthermore, the trials are still in their early stages, and the long-term efficacy and reliability need to be established through larger, independent studies.

Intriguingly, Dognosis envisions a future where the trained dogs are eventually replaced by sophisticated sensors, effectively distilling the biological prowess of a canine nose into advanced hardware. Regardless of this future evolution, the fundamental concept — leveraging the unparalleled biosensing ability of dogs, standardized by AI, to address critical gaps in cancer screening — represents a significant and compelling leap in preventative healthcare.

FAQ

Q: What is Dognosis's core technology for cancer detection?

A: Dognosis combines the acute sense of smell of trained sniffer dogs with artificial intelligence and bio-sensors. Patients breathe into a mask, which dogs then sniff. The AI interprets the dogs' reactions (movements, respiration, brain activity) to identify chemical signatures of cancer in the breath sample.

Q: What are the primary benefits of this cancer prescreening approach?

A: The main benefits include its non-invasiveness, potential low cost, and accessibility, making it suitable for widespread early detection. It can target over 20 cancer types and aims to bridge the gap in screening rates, particularly in regions where traditional diagnostics are expensive or hard to reach.

Q: What are the current limitations or caveats of the Dognosis system?

A: Dognosis is a prescreening tool, not a diagnostic, meaning a positive result prompts further medical investigation rather than a definitive diagnosis. Its reported accuracy figures are currently company-reported and require independent confirmation. Additionally, in India, it operates without regulatory approval typically required for diagnostic devices, as it's classified differently, and the technology is still in early trial phases.