We're teaching AI how to smell to solve for disease detection at scale. Our current focus is developing an ultra-affordable, non-invasive breath-based multi-cancer early detection (MCED) test. Our system combines trained dogs' exceptional olfactory abilities with brain-computer interfaces and machine learning to create quantitative signatures of disease volatilomes. Patients simply breathe into a mask for 10 minutes, which is then shipped over to our labs to be analyzed in our controlled laboratory setting.

No, we don't bring dogs directly to people. Our system uses breath samples collected in face masks, which are then presented to the trained dogs in a controlled laboratory setting. Once sealed in a bag, masks can retain potency for up to 3 months, and potentially even longer. Samples can be transported hundreds or thousands of kilometers away and sniffed by our team of highly trained dogs in a sensor-rich environment.

Actually, 40+ double-blind trials published in peer-reviewed journals have demonstrated that dogs can detect various diseases, including different types of cancer, with high accuracy. This ability is due to their extraordinary sense of smell, which can detect subtle changes in the body's volatile organic compounds (VOCs).

Our preliminary results show 98% accuracy in detecting 10 types of cancers, significantly outperforming traditional screening methods. For comparison, liquid biopsies from companies like Exact Sciences and GRAIL have high specificity but sensitivity hovers around 50%, while mammograms achieve 80-85% accuracy. Crucially, our data indicates that accuracy remains stable across both early and late stages of cancer.

The dogs are detecting changes in volatile organic compounds (VOCs) that are produced by the body when certain diseases, like cancer, are present. These VOCs create a unique "odor signature" or volatilome that trained dogs can identify, just like they can be trained to detect bombs and drugs.