7 ms·
Electronic nose for indoor mold detection and identification
- alansaber 10mo agoI was surprised to see surface acoustic wave sensing for bio mentioned, since my lecturer researched it in the 80s. I guess mechanical sensing is hard to innovate.
- andsoitis 10mo agoUsually you have to spend thousands of dollars for an expert with the right equipment to focus on select areas of your home (areas that are suspect). Are there ways to do a full-house scan yourself?
- joezydeco 10mo agoMaybe check out AirAnswers. https://www.airanswers.com/ https://www.airanswers.com/
- bflesch 10mo agoThe AirAnswers website blocks visitors from the EU. Most likely because they are in violation of GDPR and have no plan to care about data privacy. This is a really big red flag for all AirAnswers customers.
- andsoitis 10mo ago> Most likely because they are in violation of GDPR Why do you say that’s the most likely reason? You can’t imagine other plausible reasons?
- bflesch 10mo agoWhat if not GDPR would make an american B2C startup intentionally block visitors from Europe? Their investors most likely fly in and out of Europe all the time and want to show their portfolio company to their buddies. The startup's growth case is international expansion.
- ted_dunning 10mo agoBecause they don't understand GDPR and don't have time to figure out if and how it applies to them?
- bflesch 10mo agoAirAnswers is not a startup as I originally thought, it seems to be a ten year old company with mature product and a set of patents they mention in marketing materials. If they block EU customers it's definitely a conscious choice.
- Aurornis 10mo agoEverything about this website screams alternative medicine scam. "Integrative medicine" and "book a consultation" is not what you want to see on your mold testing lab's website. It means they're out to sell you things. I can almost guarantee that everybody who gets one of their test kits will get, surprise, a positive result that requires more expensive consultations and treatments from them.
- joezydeco 10mo agoOne of the founders is a friend of mine that has been an IVD researcher for decades. I don't smell any scam. They are just looking for traction in the consumer and real estate sectors. They already have customers in agriculture, especially with cannabis farms looking for molds. They don't sell treatments. It's more like a radon sampler you leave in a house for a day to get a reading.
- westurner 10mo agoWhat sort of an indoor drone could this sensor be most usefully added to?
- Ancapistani 10mo agoUAS move a lot of air - I can't imagine this would be practical. At the very least, you'd need to put the sensor on a boom long enough to make it unwieldy.
- westurner 10mo agoMany of the described robots in the mujoco_menagerie gallery could probably do the job: https://github.com/google-deepmind/mujoco_menagerie https://github.com/google-deepmind/mujoco_menagerie The sensor recovery time limits the sampling time. A 28 second sensor recovery time with SnO2-Gr (graphene oxide) would be more useful for sampling larger volumes than minutes with just SnO2 FWIU; https://news.ycombinator.com/context?id=46521695 https://news.ycombinator.com/context?id=46521695
- Aurornis 10mo agoAttaching an air sensor to a giant high-speed fan is not going to help you pinpoint anything, it's just going to blow the air around. Getting accurate room-based location requires even turning the HVAC off and avoiding a lot of walking through a building.
- gerdesj 10mo agoSwabs and a microscope. Oh and a degree in biology might help!
- bflesch 10mo agoThat's what I had to do to figure out source of my asthma and allergies. Without any microbiology degree and while suffering from delibitating disease it was very hurtful that reddit posts with requests for help identifying objects were regularly deleted from the major microscopy communities. They simply refused any discussion or assistance with DIY microscopy related to any human disease while in other subreddits people post their poop for analyis. In tune with the saying of "it can't be what mustn't be" ("Weil nicht sein kann, was nicht sein darf") a lot of medical professionals outright dispute mold-caused sicknesses. Their imaging can detect late stage fungal infections in the lungs and head of elderly people wholly consumed by the fungus, but they have no methods to detect early stage infections. And instead of realizing they're lacking appropriate analytical methods for mold detection they outright deny that it could be the cause of the problem. Luckily the microscopy helped me to figure out which samples to send to a professional in order to pin down and remediate the cause of my sickness.
- schiffern 10mo agoI gather that airborne mycotoxins also play a role in inflammation, in addition to actual fungal infections.
- bflesch 10mo agoThe airborne mycotoxin I was exposed to is a scientifically proven carcinogen and class A indoor toxin, confirmed by an official lab. It was direct cause of my eosinophilia/asthma/sinusitis. It is the biggest gaslighting ever because two people share a place and both have emergency hospital visits and nearly die but from totally different diseases in different areas of the body. All the communities suggest you throw away everything once you leave a mold-infested place, but most can't afford that. Many keep their furniture and regret it. I truly wish this experience upon the denialists so they can use it to gather the evidence they need. A proper cancer would've been easier because there is at least a support network and hospitals have a damn handbook for it.
- z2 10mo agoI've had success using my nose and a cheap $15 moisture meter against walls and ceilings. In my case I was already pretty sure the moisture came from the roof when it rained so went looking right after rainstorms. This was after also running a dehumidifier in a formerly damp basement. If a home has an active mold problem, it probably has an active water or moisture problem. What mold remediation people sometimes do as well is use an IR camera to try to find unusually cold (thus damp areas).
- tdeck 10mo agoI've seen situations where experts put different colored water soluble dye on different spots outside the house, so that when it leaks through you can determine the source. Presumably that's within reach for an individual as well.
- taneq 10mo agoThe IR camera idea is clever, usually we use them to find hot spots. Gonna try this next time I suspect we have a leak.
- Aurornis 10mo ago> If a home has an active mold problem, it probably has an active water or moisture problem. Mold can't grow or spread without moisture, so a moisture problem is a necessary prerequisite for a mold problem. So focusing on fixing any moisture problems is a great place to start. Feeling around walls and baseboards or climbing up into the attic in the hours and days after a big rainstorm is one way to get started without any equipment investment. Air circulation also helps dry things out, so make sure every space has some openings for air exchange.
- westurner 10mo agoLooks like adding Graphene to SnO2 (Tin Oxide) sensors increases sensitivity and battery life and reduces operating temperature to room temperature, for an increase in manufacturing complexity and cost.
- westurner 10mo ago"Extraordinary Improvement of Gas-Sensing Performances in SnO2 Nanofibers Due to Creation of Local p–n Heterojunctions by Loading Reduced Graphene Oxide Nanosheets" (2015) https://pubs.acs.org/doi/10.1021/am5071656 https://pubs.acs.org/doi/10.1021/am5071656 "Highly Sensitive and Selective SnO2-Gr Sensor Photoactivated for Detection of Low NO2 Concentrations at Room Temperature" (2024) https://www.mdpi.com/2079-4991/14/24/1994 https://www.mdpi.com/2079-4991/14/24/1994 ; UV photo activation of SnO2-Gr sensors "Engineering of SnO2–Graphene Oxide Nanoheterojunctions for Selective Room-Temperature Chemical Sensing and Optoelectronic Devices" (2020) https://pubs.acs.org/doi/10.1021/acsami.0c09178 https://pubs.acs.org/doi/10.1021/acsami.0c09178 "Humidity-sensing Performance of Graphene/SnO 2 Nanocomposites" (2025) https://sensors.myu-group.co.jp/sm_pdf/SM4049.pdf https://sensors.myu-group.co.jp/sm_pdf/SM4049.pdf ; 28s recovery time instead of minutes
- bflesch 10mo agoThis is significant, especially if it can be productized. Thanks for working on such an important topic. One day in the future technologies like this will keep vulnerable, sick people from being gaslit by other humans about the cause of their disease.
- HocusLocus 10mo agoOr just project negative ions into your working and sleeping places and breathe easy, wipe the mold spores off the walls every couple months or so. Test their effectiveness by shining a bright flashlight in the dark. You can't see the beam!
- taneq 10mo agoAre you saying that all particulates in the air inside a house are mold spores? Surely the vast majority is just normal dust (dander, microplastics, silicates, etc.)?
- DANmode 10mo agoDust becomes mold rapidly. Way more rapidly than anyone’s comfortable. Ambient air has mold spores. Add a single humid breeze through the space, game over. If you don’t have a humidity range being recorded day to day in your home, you may be surprised the excursions.
- adamddev1 10mo agoHow do you "project negative ions" ?
- jadamson 10mo agoWith an ionic air purifier.
- m348e912 10mo agoOzone (the source of those negative ions) comes with its own issues. If you are going to use ionize with ozone, it's best to do it when you're not going to be home for a while.
- jadamson 10mo agoOzone doesn't generate ions, ionizers produce ozone, and how much will depend on the device.
- ram_rattle 10mo agoThis e-nose in general interesting, do anyone know any sensors that are currently commercially available to prototype these things?
- pedro_caetano 10mo agoKIT had a project on this (Project KAMINA) which looks like it is also cited in the linked paper. They had a spin off SMELLDECT GmbH which sells a kit but not exactly a order from DigiKey thing. I imagine you will need to send an RFQ and go through the motions with their sales team.
- ram_rattle 10mo agointeresting, thanks for sharing these!
- sschueller 10mo agoI find it interesting that we seem to have mastered what the human eye can do and even go beyond it with like infrared but somehow we still can't build a chip that can "taste" or "smell".
- lopis 10mo agoMakes total sense to me. Detecting and measuring photons seems much simpler than accurately detecting whole molecules. When we need to detect if a sample contains a certain kind of molecule, it usually requires expensive chemical processes.
- throwaway198846 10mo agoOr Mass spectrometry
- Terr_ 10mo agoPretty sure that relies on a lot of assumptions about how the pieces being detected originally fit together, or extremely pure samples.
- Latitude7973 10mo agoMass spectrometry is a multi-faceted beast, and has many applications. One that comes to mind is ion mobility spectrometry. I am most familiar with FAIMS which is an extremely selective method and can detect trace amounts of specific molecules. The challenge is piecing together what you have detected to draw conclusions about the sample - in this instance you might detect a specific molecule, but to definitively conclude that it's caused by a particular fungus requires lots of prior testing.
- Razengan 10mo agoIf you just want a histogram of all the chemicals that are present, that would probably be doable if not already done. But how would you even quantify/qualify the "sensations" of those senses? Vision is "easy": What I see is what you see is what the machine sees. A machine shows us what it sees and we can verify that it is working correctly, with a glance. How would we verify that a machine smells or tastes "correctly"?
- tippa123 10mo agoThis would be very interesting for industrial applications that rely on clean rooms. What comes to mind is environmental monitoring for microbes, which is tedious but important and currently has significant lag time (≈2 weeks). What I couldn’t determine is whether there is any information about concentration. I mean so what if something is present, you have a lot of ubiquitous organisms out there. Nevertheless, if the technology matures, it could help identify a problem earlier before it becomes visually obvious. You would still need to determine the root cause. Or it could help with a better decision making before buying a house? All in all a lot to think about.
- npsomaratna 10mo agoWhy is there a significant time lag? Is it because monitoring is done via cultures?
- tippa123 10mo agoCorrect, the workflow is something like this usually. You would take contact, settle and active air sample plates within the cleanroom, followed by approximately one day before culturing is initiated in QC. Incubation then typically takes around seven days to cover both bacteria and fungi. You then get the colony forming units value which is the key parameter. Some companies take this further and perform organism identification, which adds additional days to the timeline but great for reactive investigations. There is also a lag until the data becomes available in a digital format. This of course differs between companies. Some companies may opt for shorter or longer incubation times, but in general, the key takeaway is that the process takes time.
- alfonsodev 10mo agoAt the risk of sounding stupid, I’ve noticed my mom’s house smell has changed a lot in the last year and is stronger in her room and bathroom, coincidentally or not, since she has started showing mild dementia symptoms and has been diagnosed with vascular dementia. She feels extremely tired despite standard blood test markers being OK. The smell could have simple explanations, like she forgets to ventilate the room, or her hygiene is not as good as before, since she has become more avoidant of any discomfort, but we are helping her with the hygiene part at least, the ventilation is harder because she complains about cold, and refuses to leave the room. Also to me, it is a different smell from what would be poor ventilation and from what would be lack of hygiene, but I can’t point out what it is. I'm usually very good detecting when a room has not been ventilated, I just feel I breath worse and can't focus as well, I'm annoying to other people that doesn't want to open the window. She spends most of the time in bed, her body is telling her to sleep much more than before, and she has become very stubborn about getting out of the house, going to the dentist, etc. Since she is very intelligent and still has the verbal ability to make you believe she will do it as soon as she feels better and that will be soon, the reality is that months are passing by and that is not happening. I wonder if anyone could recommend an affordable but effective air quality device, mold testing kits (if such a thing exists), or what targeted blood tests can be done. Also, if there are tests to fine grain the diagnosis of vascular dementia, to know if there is accumulation of proteins in the brain, or what is the origin of the issue. She is 79 years old, perhaps there is nothing to do, but somehow I feel the environment of the house has something toxic in it, I even thought about electromagnetic fields, but I really don’t know where to start. Edit: We live in Spain, her doctor feedback is that her blood tests are great for her age, and we didn't have much info about the vascular dementia situation somehow has been implied that is normal in her age.
- adastra22 10mo agoI don’t have specific suggestions for your questions, but that does sound like mold, and the mold could be progressing those symptoms.
- alfonsodev 10mo ago> but that does sound like mold, and the mold could be progressing those symptoms That's both my fear and my hope, if you have any good resource that I could educate myself about this hypothetical type of mold that can accelerate the symptoms of dementia, please share, would be very useful. Thank you
- crispyambulance 10mo agoI’ve used hardware store “mold kits” where you tack some tape onto a surface and then send it to a service that will analyze it for you. My understanding is that these services simply look at the tape through a microscope, or apply it to a growth media that’s been prepped to “prefer” the specific types of fungus they’re looking for. One would think there are PCR-based services that do this? That would be the gold standard for this stuff, and it could easily scale enough to become economical, but to my knowledge there are no commercial mold testers that do this.
- voxleone 10mo agoThe inevitable AI angle: researchers are increasingly exploring perception, self-modeling, and grounded interaction as the foundation for the next wave of AI systems, ones that behave in ways more aligned with human-like awareness than LLMs alone can provide (see work like MUSE or situational awareness in vision-language reasoning). Systems like the electronic nose described here highlight what many think is missing in current AI approaches: continuous physical sensing combined with explicit novelty detection and decision boundaries that let a system say “this is real”, “this is happening now”, or “this is outside what I know”. Human-like behavior is unlikely to emerge from language models in isolation; it appears to come from closed perception-reasoning loops that are causally coupled to the environment. Without sensory grounding, AI tends to optimize for plausibility rather than correctness, and scaling or prompting alone doesn’t seem to address that gap.
- vjanma 10mo agoLLMs are trained on text about the world, not the world itself. Olfaction is an interesting test case because it's one of the most ancient and direct sensory modalities. no symbolic abstraction layer, just molecular binding triggering pattern recognition. What's compelling about pairing e-nose hardware with transformer architectures is you get that grounded perception loop you're describing. The sensor array produces high-dimensional response patterns from real physical interactions, and the model learns to classify and reason over patterns it's never been explicitly trained on—genuine novelty detection rather than interpolation over training data. The "this is outside what I know" capability is critical for real-world deployment. A model that hallucinates a scent classification is potentially dangerous (think: fentanyl detection in law enforcement). You need calibrated uncertainty, not just a softmax score.
- xtiansimon 10mo agoAnd electronic mouth to taste water. Residents in Bethpage, NY are dealing with Grumman water contamination or “plume”. https://youtu.be/vgezHCoqiUo?si=1wn7Grt8vpAnzJ_Z https://youtu.be/vgezHCoqiUo?si=1wn7Grt8vpAnzJ_Z I live on Long Island and drink well water. I’d sure like a home monitor.