Cancer has a smell, and scientists want smartphones to 'sniff out' cancer

Every home and pocket has electronics that hear, see and feel our touch. Your smartphone is likely capable of all three. What is missing? Technology that can smell. But this may change, as detailed in episode unexplainable, weekly podcast Vox, exploring unanswered questions in science.

Production technology Смартфоновthat smell comes close to reality, says Andreas Mershin, a researcher and inventor at MIT. “I think we may have five years left, maybe a little less,” he reckoned, “to get it from where it is now to fully enter the phone. And I'm talking about [deploying] a hundred million phones."

The idea isn't to have Siri tell you when you need a shower (although that might be useful for some people). There is a larger public health mission: to replicate the incredible possibilities of canine disease detection in a more portable, accessible form to help catch insidious diseases at an early stage.

Dog nose prototype robot

Dogs can sense cancer, Parkinson's disease, malaria, and other conditions that cause changes in human body odor. There are even published research about the ability of dogs to smell COVID-19.

Scientists could train more and more dogs to help detect diseases and deploy them around the world. But such preparation is costly, difficult and time consuming. In addition, not everyone likes dogs, and not everyone would like a dog sniffing them before a basketball game or during a visit to the doctor.

On the subject: Every tenth American has diabetes: what is this disease and how to prevent it

In essence, the hope for scientists is to design a robotic nose to detect diseases the way a dog's nose can. This is the topic of today's episode of Unexplainable, which is also the third episode in seriesexploring the unknown about our feelings. (Note: This episode's version originally aired in 2021.)

One of the subtleties that is so intriguing about robotic noses is that researchers have been able to come a long way in this, although it is not yet fully understood how biological noses function. This is the main mystery of this week's episode unexplainable: How can scientists build a robotic nose if they don't understand how the sense of smell works?

Robot nose challenge –6&&& explanation

Scientists know many basic mechanics of smell. They know that there are special receptors deep in our noses that interact with the odor molecules we breathe in. And they know that these receptors send signals to the brain.

What they don't know is why everything smells the way it smells. That is, why does one odor molecule smell differently, while another similarly shaped odor molecule smells differently? This is still a mystery. (For example, in vision, there are only a few types of receptors, and the way they interact is well understood. C odor, there are 400 types, and we've only just begun to understand how they work together to discover it).

Mershin first attempted to build a robotic nose in 2007 when DARPA, the research and development agency of the US Department of Defense, announced a call. He wanted to see if someone could build a robotic nose that could detect certain molecules. DARPA has not released an exact list, but based on the list, Mershin suspects the molecules are related to smell bombs, drugs, or dead bodies.

So how to build a robotic nose? Mershin, his MIT colleague Shuguang Zhang, and their team first thought about building something that could directly detect odor molecules—something like an improved carbon dioxide detector. When any odorous molecule enters the detector, just as when carbon monoxide gas enters a CO detector, an alarm is triggered.

They used real odor receptors that were grown in cells. And they spread the receptors over the circuit board. (All sensory receptors exist to convert the outside world into electrical signals that our brains can understand. In this case, the signals will be passed on to the circuit.)

It was hoped that the receptors would give readings of all odor molecules in the vicinity. But then the team realized that there were so many molecules they needed to detect, so many smells, that the alarm was going off too often. Therefore, the team decided not to build this car.

Nano-Nose

For their next attempt, which they called the Nano-Nose, Mershin and Zhang tried something different.

While their previous version was supposed to detect individual molecules, this one presented a bigger picture - the overall smell. (Smell isn't just a combination of molecules; it's a specific way these molecules interact with a complex series of receptors in the nose. It's still a mystery how it all comes together.

The Nano-Nose recorded the pattern of electrical activity produced by the receptors in the presence of an odor. The readings were like a stock ticker. Each smell, Mershin discovered, created a pattern that the computer (i.e., the brain) could recognize.

Essentially, Mershin used machine learning (a form of artificial intelligence) to allow Nano-Nose to figure out its own way of recognizing odors. Interestingly, this is how dogs are trained to sniff disease. Mershin tuned the Nano-Nose to what he needed to recognize for the test, and then allowed the Nano-Nose to teach himself to recognize odors.

Ultimately, he showed that Nano-Nose could detect odors using real receptors and do it. quite sensitive. But the DARPA test was in an extremely controlled environment, which may be why DARPA ended the program after a few years. The real world is much more complicated.

On the subject: Scientists have found an effective cure for migraine, and you can organize therapy at home

Mershin continues his research by trying to prepare the Nano-Nose for everyday use. And if previous iterations were the size of a desktop, now Mershin says he's shrunk the technology so it can fit inside a smartphone.

The noses in our phones could help save lives

Mershin still needs to train the Nano-Nose more before he can actually smell disease in everyday situations. In the real world, there are countless smells around, which is very different from the clean, controlled environment of the DARPA test. To overcome this hurdle, Mershin needs to expose the receptors to a much larger number of odors in order to improve the algorithm. But the main proof of concept is here, and phones that smell might not be far away.

Giving phones a new way to monitor our behavior comes with some distinct challenges. Will the data "sniffed" by the phone be confidential? Can the robot's nose be hacked or available to advertisers? These are the questions we're already asking about our phones' ability to see and hear us, often with unsatisfactory answers.

The privacy implications may be worrisome, but the benefit seems clear: A pocket-sized robotic nose can help save lives. “Any of us can have a tumor that becomes malignant,” Mershin says. “If you put off six months, then sometimes it becomes a death sentence.” But if your phone notices a change in smell, it may alert you sooner.

Claire Guest, a scientist who trains dogs to detect diseases, says she takes this lifesaving potential personally. Years after she began her research, her dog Daisy began to look upset. The dog constantly "stared" at Gest and pushed her somewhere. She felt where Daisy was pushing and found the lump. She was eventually diagnosed with breast cancer and it is possible that Daisy saved her owner's life.

Not everyone has access to dogs like Guest. But almost everyone has a smartphone. Right now, our phones don't smell cancer. But one day they might save a life.