Mould contamination and the use of AI [artificial intelligence] technology

Apr 27, 2023

Two years ago on June 14, 2021, I published my weekly newsletter on LinkedIn for a series called “Design healthy environments”. The article had to do with the popularity of mould-sniffing dogs and the title was “The preoccupation with mould spores is ridiculous and counterproductive”. The point I tried to make was that identifying the presence of mould was not as important as knowing what to do about the contamination. Having mould in a home is very common. Approximately 40% of homes have a mould contamination problem and most people in good health can readily detect the smell of mildew. It’s knowing how to get rid of the problem that is the key. You have to be able to get to the source, to get a proper diagnosis. It’s not just a straightforward case of cleaning up. Most so-called “cleanup experts” use dehumidifiers which in themselves become contaminated and accelerate the spread of mould spores throughout the premises. Subsequently, a strong disinfectant such as an antimicrobial product is used for the clean-up/decontamination and that chemical in itself is unhealthy. Many are hormone disrupters.

The right way to proceed according to the “NY protocol” [Guidelines on assessment and remediation of fungi in indoor environments NY City Dept of Health and Mental Hygiene?Nov 2008] is to first remove all wet material [within 48 hours], to then resolve the source of the problem before starting to rebuild. I also mentioned the need to check for all possible interior contaminants, not only mould spores but include all particles less than 2.5 μm [microns] which can get into your lungs and subsequently your bloodstream. It would include bacteria and viruses.

I didn’t realize that criticizing dogs would offend many dog lovers and most of the comments I received missed the point that I was trying to make – that you need an expert to do a diagnosis, to know how to create a healthy environment. In Canada, it means having a complete air change with filtered outside air, the year-round and that air change has to be accomplished within 3 to 4 hours and distributed to all the inhabited areas including basements which typically get contaminated because of unexpected flooding or negative drainage around the house.

The other day, I listened to a CBC interview regarding research that was being done at the San Diego Zoo where they were training large Australian rats whose sniffing abilities were even more remarkable than those of dogs. Apparently, they could be particularly useful in sniffing out buried explosives in minefields or roadside bombs. Certainly, dogs are currently being used to detect drugs at border crossings. In my article, I had no intention of minimizing the usefulness of dogs in regard to their sense of smell.

The same could be said of AI mould detection. Supposedly, the new software in that regard is based on having collected and identified over 100,000 mould images. Building inspectors are then able to send in photos of samples taken at a particular location which the computer can then correlate or match up with the images of different types of mould. It reminds me of the apps now available which can identify different plants and recommend what to do in case they appear diseased. I've tried it and it doesn't work so well.

The limitation of using AI for mould detection is similar to what I have written about mould-sniffing dogs. You still need a building expert [architect or engineer] to do a diagnosis, to tell you how to resolve the problem at its source [a building defect] in a sustainable manner [so it does not reoccur]. There is a parallel in the medical field which is more advanced than the construction industry. You may suffer a bad fall and you go to the ER at a hospital to get X-rayed. However, it takes a specialist like a radiologist to examine the X-rays, to identify what damages have occurred and then you have to work with an orthopaedic specialist who considers interconnected bones, ligaments, and muscles; a cast may be necessary as part of your cure to get you back to good health. AI cannot substitute for that type of on-site diagnosis based on a holistic [the parts of a building are interconnected and can be explained only by reference to the whole] inspection by an architect or engineer.

#mouldremediation #mouldsniffingdogs #AImouldidentification #diagnosis #buildinginspection #NYprotocol #createhealthyenvironments?#freshairchanges #architects #engineers #radiologists #holisticsustainable