Virus detection

Projects and Ideas
#1

Would it be possible to make a virus detector for use at say the entrance to supermarkets etc?
Initial thought would be to use a laser and photodetector. To detect scatter from particles.
The person would blow through an orifice, preferably without touching it though! The laser and photodetector would be mounted at right angles, so that the breath is checked for viruses by light scattering from them.
This technique is used for detecting bacteria:


This article describes use of a reflective sphere, but not sure how that would work in practice. A reflective tube would be more practical. For example an aluminium can with the ends removed.
Also, it could be made insensitive to ambient light by driving the laser with a squarewave, and synchronously detecting (lock in amplifier technique) the photodetector output.
It would not be able to distinguish different types of virus. And also something would need to be done to avoid false alarms from particles such as pollen and exhaust fumes.
Does anyone have some spare time to have a go? I need to focus on the oxygen generator for now.

Any comments?

#2

My guess is that the virus is too small for optical detection. But it may be that it’s spread mostly in droplets, from coughs and sneezes, and those droplets are larger. The result then would be detecting coughs and sneezes - which might or might not be laden with virus. My suspicion is that people are now quite alert to coughs and sneezes, and my observation is that people are trying not to get too close, or remain too close, to each other. Although I don’t think we yet have big red marks on the floor to illustrate a 2m distance for queuing, I’ve seen a photo of such, in Denmark.

#3

As I understand it, people can show few symptoms initially, (i.e. not yet coughing) and yet already be breathing out viruses.
Is it just a matter of going to shorter wavelengths? Maybe using UV LEDs would do, rather than a laser?
Would a differential approach help discriminate against pollen and smoke particles? Say have a red or IR LED for detecting these, which would be less sensitive to the virus, and at the same time (or interleaved timing) pulse the UV LED for detecting virus. The latter would also detect pollen etc, but then subtract one measurement from the other to remove the measurement of pollen etc.

#4

Even UV seems to have a wavelength probably twice the size of the virus: 200nm vs 100nm, in round numbers.

As I understand it, if you’re in a high risk group, or caring for someone in such, you do need to do what you can to avoid infection. That would, I think, include not going into shops and probably also not walking down the high street.

Otherwise, if you’re in the majority, not at high risk of a bad outcome, you need to do what you can to avoid transmission - which is a slightly different position.

What we do as individuals affects the statistics, and it’s the statistics at large scale which determine how this epidemic evolves. Any one of us could choose to be extra-careful, or extra-carefree, and it wouldn’t make much difference, so long as it’s not too many. (Edit: actually, this might be a terrible view and a poor summary. Seek professional advice!)

#5

" Each SARS-CoV-2 virion is approximately 50–200 nanometres in diameter." - link

The lowest size I get from the sensors I have is 0.3 micrometers (μm), so 300nm, just out of range, but would be in the UV range. If the UV source works, then the numbers from the conventional PM detectors could be used as a cross-check to see whether other particles are triggering it.

#6

Will,
Does the sensor actually detect smaller particles, albeit at a reduced sensitivity?
What wavelength does the sensor use?

#7

Alternatively, using two sensors, one modified and the other unmodified.
The modified one using a UV LED. Passing the air from one to another so that they are measuring the same thing.
Then subtract the readings to make it overall relatively insensitive to pollen and smoke particles.

#8

I can’t find the wavelength, but this is the device I have at home:
https://wiki.liutyi.info/display/CO2/D9+Inside


Techshed devices us: SDS011

+ a couple of papers on the theory/evaluation of the devices (a bit of a rabbit hole):
https://uwspace.uwaterloo.ca/bitstream/handle/10012/12776/Tan_Ben.pdf?sequence=5&isAllowed=y

#9

Thinking about this a bit further - it’s said to be “droplet borne” rather than truely airborne, so the majority will clumped in water droplets of varying size (larger droplets hit surfaces pretty fast, hence the focus on surfaces and hands).

Given the particles which remain airborne are in water droplets, I’m not sure we’d be able to differentiate “normal” human droplets from those containing SAR-CoV2 particles in air…

#10

I can usually smell when someone has a cold, and I get well away from them. So if there is a detectable difference in exhalation, then can we make a detector for it? My suggestion of using light scattering was just an idea. There may be better ways to do it?

#11

It’s a good idea, I just haven’t seen what tools could be used.

I haven’t found much useful yet - probably needs to be chemical detection - testing seems to be DNA/RNA test (PCR) at this point, but you’d think there would be almost a “honeypot” chemical to allow things to bind, then…
[interrupted b by the thought this is like looking for an antiviral]
… It’s a bit less strict than making a medicine - this just needs to change in response to interaction with the virus particles.

An alternative approach might be early signs of infection changing body odour as a side effect of immune responses, but that feels like a harder problem given the amount of variation in baseline there might be…