Dogs’ sensitive noses and their ability to be trained to detect a variety of medical issues/diseases are well known. Using their sense of smell, trained dogs can detect lung, breast, ovarian, bladder and prostate cancers. For prostate cancer, dogs had a 99 percent success rate in detecting the disease in patients simply by sniffing urine samples. In fact, many different types of cancer have been detected earlier by dogs than any other technology.

For more than 15 years, researchers have been pursuing a way to emulate this capability with manufacturable technology. In fact, a team of researchers at the Massachusetts Institute of Technology (MIT) developed and continues to improve on a patented detector system that incorporates mammalian olfactory receptors stabilized to act as sensors. The sensor system can detect the chemical and microbial content of an air sample with even greater sensitivity than a dog’s nose. The problem is connecting the dots. Different cancer samples that a dog can detect have nothing in common. Yet a trained dog can generalize from one type of cancer and identify others.

To address the problem, the researchers added a machine-learning process with an artificial neural network (ANN) to identify the distinctive characteristics of the disease-bearing samples sensed by their electronic nose. Using powerful analytical tools including gas chromatography mass spectrometry (GCMS) and microbial profiling, the artificial system was able to match the success rates of the dogs, with both methods scoring more than 70 percent accurate detection. Confirmed through DARPA-mandated control tests, the miniaturized detection system is 200 times more sensitive than a dog’s nose in its ability to detect and identify tiny traces of different molecules.

Since the data streams can be handled in real-time by a typical smartphone, researchers envision an automated odor-detection system small enough to be incorporated into future smartphones.