Love seeing these types of posts on HN! Every now and then I see a post about olfaction, and since I am in the olfactory receptor research space, I just want to put my two cents out there.
Digital chemical sensing is hard, and yet biological chemical sensing is amazing. Understandably, there's always been immense interest in digitizing the biological process of smell. People have been pitching various technologies, from the early days of the WaspHound[0] and DARPA's largely failed RealNose project[1], to current efforts from startups[2] and big players in perfumery[3].
It is my opinion that all of these technologies are solving the wrong problem. Olfactory receptors are just plain horrible proteins. They're extremely difficult to produce, and even when you can express them, they have exactly the wrong type of properties. Each olfactory receptor responds to many different chemicals with really bad sensitivity. While that works really well for animals, these properties are a nightmare for digitization.
If we want to digitize chemical sensing, I believe we will need better proteins than naturally occurring olfactory receptors. And that's what I am trying to do currently in my Ph.D. research!
If you haven't read "What the fly’s nose tells the fly’s brain" [0] take a look. It helps explain why the olfactory sensors might actually be selected to have wide spectrum sensitivity (it's a feature not a bug).
Thanks for the link! Cool to think about how the evolution of smell has optimized for "a maximum entropy code."
I totally agree, it's a feature not a bug. A wide spectrum is absolutely an essential feature for an animal to sense thousands (to millions) of different chemicals without thousands (to millions) of different receptors. So, calling olfactory receptors "just plain horrible proteins" was a bit rude to them, haha!
However, in my opinion, digital sensors based on olfactory receptors have and will continue to suffer from such wide spectra and poor sensitivities.
It's interesting to think of this from a security point of view. What data can be exfiltrated? How much signal are we giving off all the time that would leak information on way more of what's going on inside than we realize?
If Spectre can use a timing side channel to peer into regions of memory that were not supposed to be accessed, how many more side channels are there in a human body that mostly wasn't designed to conceal anything? (That's not exactly true, we've evolved to hide quietly, many things hunt by smell, women conceal their state of fertility, etc. But technology can access a much lower threshold of signal than anything evolution has had to worry about.)
If you listen closely enough, I'd expect we're all open books. Time the exact duration that nerve impulses travel, that blood coagulates, that sweat is produced, that hormones are released, etc. Look at the ratios of chemicals produced by digestion, respiration, locomotion, etc.
Technologists are already mining some of this wealth of signal from just filtered speech recordings. There's so much more that could be harnessed (and abused).
Technically, this doesn't seem to be a nose per se, but more creating a sensor to detect molecules for which the mammalian nose has evolved to detect. If we have an excitatory pattern, then most AI techniques should be able to detect it easily.
Instead of creating a single modality i.e. artificial noses or ears, medicine would benefit from a multimodal sensory system that uses vision, smell, auditory and perhaps even taste, _in conjunction_, as the human body is rich with signals through all these modalities.
Given the decades of research into an electronic sense of smell and the billions of $ it will be worth if we can get a electronic nose, why are we jumping the gun to the other things?
> a sensor to detect molecules for which the mammalian nose has evolved to detect.
This would be a modern marvel and so far impossible. It also wouldn't have to be mammalian, there's no reason to limit it.
> If we have an excitatory pattern, then most AI techniques should be able to detect it easily.
Exactly. I'm not sure why you'd need AI at all. It should just be a look up table.
An electronic nose is a big deal.
We are talking the ability to know what people or vehicles are carrying without a search, tracking people and animals over land, the ability to find every truffle in a field, it should help us create smells like 4D cinema, find pollution in water and air and land, track leaks, cooking, find animal and mineral resources, animal husbandry and medicine.
Yes. The sensors are the problem. Processing the data probably isn't a huge job.
There was a startup trying to do this in 1999.[1] Cyrano Sciences, Inc.
They got some Small Business Innovation grants.[2] They were trying to build a chemical weapons detector badge, an oil/fuel leak detector, and related devices. They developed a handheld unit, the Cyranose 320.[3]
They were acquired by Smiths Detection (UK). They sell various devices for detecting explosives, drugs, disease carriers, and such. "Currently trialing rapid detection airborne COVID-19."[4]
Perhaps all those kinds of sensors could be cleverly combined in ways that make them more feasible/reliable, not less.
If we limit ourselves to designing it with the same specific conceptions of smell that we have, we might be unnecessarily making the job harder for ourselves, or missing easy optimizations that would make it more effective.
How about using an animal's nose instead of an electronic nose? Detect the brain signals and let the "AI" figure out the patterns when the animal inhales something. No need to create a new nose from scratch, no need to train the animal as well.
> Instead of creating a single modality i.e. artificial noses or ears, medicine would benefit from a multimodal sensory system that uses vision, smell, auditory and perhaps even taste, _in conjunction_, as the human body is rich with signals through all these modalities.
Thats a great thought right there. Adding an over-time perspective would make this even more powerful, as a lot of information is encapsulated in change, not just absolute observation.
I wonder how much of the required sensing tech is already available, but most likely not rated for medical usage yet.
There was a cat named Oscar in the Steere House nursing home in Providence, RI, who was kind of a jerk. However, he had one incredible power. The advanced dementia ward where he lived had numerous sick patients. Sometimes, He would jump onto one of their beds. Sure enough, hours later, that patient would die. He had a remarkable record, over 20 cases in a row were predicted correctly.
It must've been smell, he sensed the deterioration from biochemical signals. Why he liked to be around death, I don't know, cats are strange!
Sounds apocryphal or just flat wrong. A cat that jumps up on a bunch of beds in such a place is bound to do so on the beds of dying patients. The dying ones probably got more visits from doctors, family, etc. and the cat could've been responding to the presence of humans, or any number of other factors. To assume it was smelling imminent death and jumping on the bed to signal it is way beyond what you could possibly prove.
It's not flat wrong nor apocryphal, check your facts, his performance was studied and published in a book and in the New England Journal -- do you think the NEJM is likely to publish pure aprocryphal falsehood?
I volunteered in the Steere House and met the cat, and my close family member was a higher-up in the org and personally witnessed the cat's abilities numerous times.
It may be true. But also: it may be coincidence, there is a chance that the cat naps on the beds of people that are lying quietly, which is something that probably precedes death more often than it does not, the cat may sleep on other beds but nobody thinks much of it and so on.
What the NEJM publishes is not always accurate or even true, there have been numerous examples of retracted studies so no need for the appeal to authority.
If it is true then someone should figure out the mechanism and prove it otherwise it might just as well be coincidence.
It is likely both. The link you provided is not a study. How many times did the cat jump up on the bed of a non-dying patient? How often on the beds of any random patient, making his jumps on the dying patients' beds simply part of the noise? Without a real study, we can't attribute superpowers to a cat.
You think it's a superpower that an animal with an extremely sensitive nose can smell that a biological organism is emitting smells of decay and death? Keep in mind, death is a process and these are advanced dementia patients, they DO have a smell as I have volunteered in the building and met many of the patients (before COVID). This is not new to science, there are human individuals world-wide who can smell diseases, for instance the Scottish woman who a physician proved could smell Parkinson's on testing her sniffing of numerous T-shirts. Biological processes can emit smells, it's really not a superpower.
By the way, in medicine, you don't need to do a randomized controlled trial to be considered a "real study". Do you think the study that found out that HIV was killing gay men in San Fransisco, a case-series study without any trial or randomization, was not a "real study"? Case series and observations are valid real studies, though not gold standards for establishing broad truths.
If you get 20 Oscars and 20 normal cats, feel free to do the RCT, or you can take a moment and do some simple Googling before going full-on skpetic on a situation you don't know anything about.
What is funny is someone killed Oscar with a bed pan after the original story come out. Sure it also is probably not true (The story also sounds apocryphal), but we can pick our realities -
Tech is interesting, title is clickbait. There are many important conditions which will leave no trace outside the body. This is why long term I’m excited about ultrasound getting cheaper, allowing individuals to look inside themselves more frequently.
As a novel signal or biomarker, figuring out aromatic traces of diseases we didn’t even know about seems cool too. So in that sense this is interesting.
But no, diagnosing disease is obviously multimodal. And I’m not at all convinced this would even be the most important modality, even in the distant future.
I'm excited about the prospect of more consumer level medical diagnosis systems. As you say cheaper ultrasound at home would be awesome, and the results could be sent to a trained physician/ai to tell you if the dark mass is supposed to be there. Putting people's health into their own hands is a positive IMHO.
Ahh, I bet this tech will be first used to add one more layer for Big Brother surveillance, and maybe if there is something else to be squeezed from it will be used for medical purposes as the article envision. Hold on to your horses because it's gonna get worse first without knowing if will ever be better after.
"Riad Sarkis, a surgeon and researcher at Saint Joseph University in Beirut, is part of a French–Lebanese project that has trained 18 dogs. Sarkis used the best two performers for the airport trial in Lebanon. The dogs screened 1,680 passengers and found 158 COVID-19 cases that were confirmed by PCR tests. The animals correctly identified negative results with 100% accuracy, and correctly detected 92% of positive cases, according to unpublished results."
btw, 10% of passengers are covid positive, and i'd guess most are asymptomatic as most of them wouldn't be flying otherwise.
From the article, it sounds like doing it in a way that approaches the sense of smell is quite hard, and no adequate approach has yet been figured out.
Electronic noses are a recurrent thought of mine. Aplications would be countless. Now, the prospect of using mice noses and then scan their brains as a shortcut feels wrong, like, animal slavery wrong. I know, these are lab mices, and we are eating millions of animals everyday, and so on, still, can't avoid the feeling. Imagine the device; air flows to a chamber filled with electrode implanted mice, the software analizes the signal and prints, in paper, a results reports. Rick & Morty league.
Combine an electronic nose with Boston Dynamics hound and you have a pretty good recipe for a nightmare or a horror movie. During the day it recharges... during the night it hunts... and it never stops chasing you.
Unfortunately, diagnosis is almost never the actual problem in any of these diseases... it's what to do when you discover it. Not to mention the problem of over-diagnosis and iatrogenic harms that happens when we know that you might have something but not where or to what extent you have it, requiring more tests and stress.
The treatment of some of these diseases (especially cancer) depends heavily on when you discover it. If this were built into every cell phone, nobody would make it past stage 1 cancer.
...well, almost nobody. "You're gonna die because you didn't call your mother often enough."
Treatment != Outcome. Until an RCT is completed, we won't actually know if such diagnostic devices are a net good or harm. You are ignoring the very real effects of iatrogenic harms and the psychological harms of giving people alerts that "cancer is detected!!!" when in actuality they would've been fine and never noticed any symptoms whatsoever. Not to mention the associated medical costs, tests done, time wasted, etc. This is why mammograms are so controversial: they might save you from dying of cancer, but they don't actually reduce your overall mortality. Screening doesn't always save lives by default. It's complicated. https://www.bmj.com/content/352/bmj.h6080
Sure. I'm aware of the difficult tradeoffs of screening. But the fact that screening tests X and Y are not worth doing isn't really adequate to write off a whole new world of additional screening tests, many of which would have very different tradeoffs. (Olfactory tests are far less invasive, for one.)
It's also something of a UX issue. Nobody is going to write an app that rings the cancer alarm after one positive. People might write an app that sends raw results to your doctor or medical monitoring service, and you get an email requesting an appointment to discuss your latest status. And if your breath results say that you're gonna die in the next 10 days and there's nothing to be done (exaggerated for effect), well then... maybe it won't send an email after all.
I honestly think the aversion to screening has more to do with other political and economic factors. For example, lung cancer screening has been shown to clearly reduce all-cause mortality in an extremely comprehensive study with 7 years of followup (NLST). These results are also replicated in studies around the world.
The problem is, how do you properly incentivize patients, providers and payers to actually implement this? The payoffs are not immediate but are rather long term for everyone involved: it takes 6+ years to clearly see the significant mortality reduction payoff actually show up in the data. The downsides, on the other hand, are immediate (another trip to the hospital/imaging center, deaths due to false positives, increased costs, more work for clinicians).
Incentivising patients should be the easiest part. People, generally, would like to live longer, and if you make imaging convenient and work on the messaging, you can get good compliance, at least for the screening part; compliance for interventions would still be an issue, I presume.
For providers, provide the studies and incentives. Bonuses for lung cancer screening work, or % of patients screened (of eligible patients seen above age X, who haven't been screened, don't have a counter indication to screening, etc).
For payers, I think there's two classes of payers; long term payers like national health services; and short term payers like US health insurance where people can change insurance every year.
Long term insurers can (and should) be looking at their budgets and data (their own, but also studies like you mention) to find out what things to prioritize. Depening on your perspective, goals could be to maximize good years of life given the budget or to minimize future budget needs given the current budget.
Short term payers are going to have a hard time prioritizing long term issues. Reducing future costs for your insurance pool doesn't help you much when your pool can change every year. To incentivize this through economics, we would need to somehow increase the time exposure of the provider. Something along the lines of the insurance carried for a given year has partial responsibility for future years. Of course, that would be an administrative nightmare.
I don't know the literature as well for Breast Cancer, but at least as implemented in the UK there seems to be clear evidence for reduction in all-cause mortality. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6121088/
Edit: hastily posted an article without reading it... this is not correct.
This study demonstrates what would be necessary to show a possible reduction - it doesn't show an actual reduction. But it would be encouraging to see such a study at some point.
Also I think you are painting an ongoing disagreement in the medical community (do we need to measure all-cause, or just breast cancer related mortality) as settled.
There are good arguments on both sides, but I think that we have a much better shot at turning the breast-cancer specific mortality boost into all-cause mortality boost by continuing to improve treatment, biopsy, screening etc.
Yeah I messed this one up. Thanks. I think my point about Lung Cancer in the other thread. still stands.
There are no RCTs, but at this point it seems not likely to happen for mammography. Perhaps for DBT though. I am optimistic that DBT + AI assistance will produce significantly different results than screening using a single reader on mammography.
Digital chemical sensing is hard, and yet biological chemical sensing is amazing. Understandably, there's always been immense interest in digitizing the biological process of smell. People have been pitching various technologies, from the early days of the WaspHound[0] and DARPA's largely failed RealNose project[1], to current efforts from startups[2] and big players in perfumery[3].
It is my opinion that all of these technologies are solving the wrong problem. Olfactory receptors are just plain horrible proteins. They're extremely difficult to produce, and even when you can express them, they have exactly the wrong type of properties. Each olfactory receptor responds to many different chemicals with really bad sensitivity. While that works really well for animals, these properties are a nightmare for digitization.
If we want to digitize chemical sensing, I believe we will need better proteins than naturally occurring olfactory receptors. And that's what I am trying to do currently in my Ph.D. research!
[0] https://en.wikipedia.org/wiki/Hymenoptera_training [1] https://www.defensedaily.com/darpa-awards-contracts-for-sens... [2] https://yesse.tech/ [3] https://www.firmenich.com/company/research