Disease-sniffing device that rivals a dog’s nose

Image credit : Bechewy

But it takes time to coach such dogs, and their availability and time is limited. Scientists are looking for ways of automating the amazing olfactory capabilities of the canine nose and brain, during a compact device. Now, a team of researchers at MIT and other institutions has come up with a system which will detect the chemical and microbial content of an air sample with even greater sensitivity than a dog’s nose. They coupled this to a machine-learning process which will identify the distinctive characteristics of the disease-bearing samples.

The findings, which the researchers say could someday cause an automatic odor-detection system sufficiently small to be incorporated into a cellphone, are being published today within the journal PLOS One, during a paper by Clare Guest of Medical Detection Dogs in the U.K., Research Scientist Andreas Mershin of MIT, and 18 others at Johns Hopkins University, the prostatic adenocarcinoma Foundation, and a number of other other universities and organizations.

“Dogs, for now 15 years approximately , are shown to be the earliest, most accurate disease detectors for anything that we’ve ever tried,” Mershin says. And their performance in controlled tests has in some cases exceeded that of the simplest current lab tests, he says. “So far, many various kinds of cancer are detected earlier by dogs than the other technology.”

What’s more, the dogs apparently obtain connections that have thus far eluded human researchers: When trained to respond to samples from patients with one sort of cancer, some dogs have then identified several other sorts of cancer — although the similarities between the samples weren’t evident to humans.

These dogs can identify “cancers that do not have any identical biomolecular signatures in common, nothing within the odorants,” Mershin says. Using powerful analytical tools including gas chromatography mass spectrometry (GCMS) and microbial profiling, “if you analyze the samples from, let’s say, carcinoma and bladder cancer and lung cancer and breast cancer — all things that the dog has been shown to be ready to detect — they need nothing in common.” Yet the dog can somehow generalize from one kind of cancer to be ready to identify the others.

Mershin and therefore the team over the previous couple of years have developed, and continued to enhance on, a miniaturized detector system that comes with mammalian olfactory receptors stabilized to act as sensors, whose data streams are often handled in real-time by a typical smartphone’s capabilities. He envisions each day when every phone will have a scent detector inbuilt , even as cameras are now ubiquitous in phones. Such detectors, equipped with advanced algorithms developed through machine learning, could potentially acquire early signs of disease far before typical screening regimes, he says — and will even warn of smoke or a gas leak also .

In the latest tests, the team tested 50 samples of urine from confirmed cases of prostate cancer and controls known to be freed from the disease, using both dogs trained and handled by Medical Detection Dogs in the U.K. and therefore the miniaturized detection system. They then applied a machine-learning program to tease out any similarities and differences between the samples that would help the sensor-based system to spot the disease. In testing an equivalent samples, the synthetic system was ready to match the success rates of the dogs, with both methods scoring quite 70 percent.

The miniaturized detection system, Mershin says, is really 200 times more sensitive than a dog’s nose in terms of having the ability to detect and identify tiny traces of various molecules, as confirmed through controlled tests mandated by DARPA. But in terms of interpreting those molecules, “it’s 100% dumber.” That’s where the machine learning comes in, to undertake for finding the elusive patterns that dogs can infer from the scent, but humans haven’t been ready to grasp from a qualitative analysis .

“The dogs do not know any chemistry,” Mershin says. “They don’t see an inventory of molecules appear in their head. once you smell a cup of coffee, you do not see an inventory of names and concentrations, you are feeling an integrated sensation. That sensation of scent character is what the dogs can mine.”

While the physical apparatus for detecting and analyzing the molecules in air has been under development for several years, with much of the main target on reducing its size, so far the analysis was lacking. “We knew that the sensors are already better than what the dogs can neutralize terms of the limit of detection, but what we’ve not shown before is that we will train a man-made intelligence to mimic the dogs,” he says. “And now we’ve shown that we will do that . We’ve shown that what the dog does are often replicated to a particular extent.”

This achievement, the researchers say, provides a solid framework for further research to develop the technology to A level suitable for clinical use. Mershin hopes to be ready to test a far larger set of samples, perhaps 5,000, to pinpoint in greater detail the many indicators of disease. But such testing doesn’t come cheap: It costs about $1,000 per sample for clinically tested and authorized samples of disease-carrying and disease-free urine to be collected, documented, shipped, and analyzed he says.

Reflecting on how he became involved during this research, Mershin recalled a study of bladder cancer detection, during which a dog kept misidentifying one member of the control group as being positive for the disease, albeit he had been specifically selected supported hospital tests as being disease free. The patient, who knew about the dog’s test, opted to possess further tests, and a couple of months later was found to possess the disease at a really early stage. “Even though it’s only one case, I even have to admit that did sway me,” Mershin says.

The team included researchers at MIT, Johns Hopkins University in Maryland, Medical Detection Dogs in Milton Keynes, U.K., the Cambridge Polymer Group, the prostatic adenocarcinoma Foundation, the University of Texas at El Paso , Imagination Engines, and Harvard University . The research was supported by the prostate cancer Foundation, the National Cancer Institute, and also the National Institutes of Health.

Source: Materials provided by Massachusetts Institute of Technology. Original written by David L. Chandler. Note: Content may be edited for style and length.