This is not rapid, only 10 tests/day, whilst the current overly slow sequencers can do 200/day. The accuracy of 5% is also worse than other faster sequencers of 3%.
And compare that to modern ThermalFischer parallel PCR sequencers, which operate in realtime, 25 tests in parallel, for 2.500€. That's a factor of 1000. With 10 modern sequencers per city you could test 10.000 per day, eg with drive-in.
Agreed. Also the Cepheid one is faster (45 minute for point-of-care results) and has FDA approval ... not sure other advantages/disadvantages of each system though (instrument and reagent cost, # deployed units, etc).
I don't work in an area where my skills can contribute but I'm in the process of finding out if I can sort insurance out to go out and do medication deliveries on motorcycle for local pharmacises, with the lockdown and the number of vulnerable people (including my mother and best friend) who can't leave their houses there is >1.5m people who can't leave at all for health reasons most of whom require prescription medication monthly.
At the moment I feel so incredibly fucking useless lurking at home playing video games.
Every day you don't fill a hospital bed can save countless lives. From the doctors who would have to diagnose and treat you to the people that you would infect when you don't realize you have the virus.
If you’re healthy you could also reach out to a local shelter, neighborhood not for profit, charity or church. They may have needs.
Also, don’t forget to socialize in whatever way you can. Skype your relatives, call your friends, send a letter to your neighbor. Go on a walk and ask people (from 6 Ft away or more of course) how they’re doing. Take time to do the human things. Anyway to make someone else feel slightly normal is this time is a big help!
In addition to all information above this test is able to detect antibodies only not virus. So it is not recommended tests by WHO. This test is only useful in the later part of infection by the time which patient might have in contact with others thus spreading the disease. We need tests which can detect single of very few virus particle in the early stage of infection and that will help to quarantine the patient thus reducing virus spread.
...able to detect antibodies only not virus...This test is only useful in the later part of infection
Actually, getting a sense of the penetration of infection into the general population, among asymptomatic people and mild cases would be very valuable information! Antibody tests could be used to sample test the public at large.
I'm not sure this is true. Bosh's site says "Which detection method is used in Bosch’s COVID-19 test? The COVID-19 test is based on a combination of sample preparation (including process controls): Multiplex PCR (Polymerase-Chain-Reaction), μArray-detection to allow the identification of SARS-CoV-2."
Even if you have unlimited virus tests, antibody tests are extremely useful. The one combined with the other demonstrates recovery as opposed to pre-infection. Supposedly antibody tests can also be made field-deployable much easier than virus tests.
I believe that line in the Q&A is referring to the "other COVID-19 tests, where the test result is available in a few minutes". It's ambiguous and confusingly phrased, but it seems like Bosch directly detecting the pathogen, in contrast to other rapid tests (which don't, presumably, use PCR, as that takes longer than "a few minutes").
I'd assume most general practitioners aren't taking patients in person anymore except in exceptional circumstances. It could definitely be useful for testing hospital personnel before they go on shift, tho.
No, this is faster than e.g. the Roche tests mentioned here last week, it will give you the first test result after two and a half hours, while the Roche one will take three and a half hours to give you the first result. So it is faster.
But it will give you the second test result after five hours, while the Roche machine will give you the second result after three and a half hours, too. As well as the third, the fourth, and the fifth. Oh, and the onehundredandninetysecond one, too. In the 24 hours the Bosch machine produces ten results, the Roche one will produce more than fourthousand results.
Tell us more. Another comment has a link to a news article that says this Bosch machine costs 15.000€ with cartridges costing 50-100€. The Bosch machine has a faster response (2.5hr), but lower overall throughput.
How much does the ThermoFischer sequencer cost? Is it affordable enough that every hospital just buys one and becomes a regional testing lab? How much specialized training does it need? (in every major city there are tons of biology and biology-adjacent majors who've done lab work but discovered there's no living to be made in bio so have switched to other fields).
Example of low latency connection: Morse code via flashlight.
Example of high bandwidth connection: van full of books.
These are orthogonal properties and must be considered separately.
The massively parallel machines will give you 200 samples per day, but in the very best case you have to wait until the end of the day to get the first one. This technology gives you a result a couple of hours after taking the sample.
Honest question, are you speaking from experience? We run PCRs at our company for our vaccine research and having to run them back-to-back day-in-day out, without any manual errors, is going to be super stressful for any lab. Not to mention the costs of training healthcare personnel on how to run a PCR.
This test is made for use at the point of care, apparently all the operater has to do is insert a swab into a cartridge and put it into the device. 
Comparing the throughput to fully automated systems used in testing labs  (look at this beast!) or qPCR machines in research labs  is missing the point, those require high investment or skilled staff. A test that can be run quickly by any doctor would be very useful to rapidly detect new outbreaks in the time following the current lockdowns.
However I suspect that the downside of this device will be the high cost of consumables, as is common for microarray biochip based devices.
The full quote is "A Bosch Vivalytic analyzer can perform up to ten tests in the space of 24 hours. This means it takes just 100 devices to evaluate up to 1,000 tests per day." which makes more sense, the 100 devices is just an example.
But I agree that it's largely unnecessary, the initial "up to ten tests per day" would have been enough.
I'm not sure how they price these devices but it could make sense to compare against labs. The question is not how many tests a device can perform but how many tests you can perform for all available devices and at a price comparable to larger lab installations. Maybe 100 devices is a comparable rate there.
True. This is seems to be on one of the extreme ends of latency versus throughput optimization and its niche in the big picture of Covid-19 will be very small.
But it appears to be a new test cartridge for an already deployed machine, so it makes sense to have it anyways. I don't know who the typical customer of an on-site lab in a box like that is, but I guess that as last part of the market will be defined by remoteness and there it absolutely makes sense.
I don't know how many of these machines are already deployed, but if there are a lot of them already at many doctors, it could be a huge thing, as suddenly, the tests can be run on existing infrastructure.
Don't current testing machines have much, much higher throughput? Like in the thousands of tests per day? It seems the advantage of this new device is latency. Of course if they are cheap enough and easy to use that you can buy thousands of them and distribute to all hospitals then might also increase total throughput.
That describes how this particular test works. It's a small device with self-contained cartridges, so it's designed to be used outside a lab and doesn't require experts to operate. But as far as I can see, it only processes a single sample at a time. A typical PCR setup in the lab can handle 96 reactions at a time, though you also need controls there which I assume are built in the cartridge in some way in this test. And then there are the big automated systems.
I think the benefit of this device here is only present if you already have one, it's not a way to seriously scale up testing as a whole unless these devices are very widely used already.
The idea is to not have a lab with a huge machine testing several thousand samples a day, but to have these devices spread out throughout the hospital and doctor's offices to test closer to the patient. A device will also be much cheaper than the fully automated devices in labs.
If it works and they can provide enough devices, this could be great. Logistics (getting samples and material to labs and handling them there) is one of the constraining factors at the moment.
It’s a really small device as far as I understand, so you could use it in a small practice or even on the go (?). This can be an advantage as most of the lab systems weigh several tons and take up a lot of floor space, hence you can never use them in a decentralized setting.
The intersection between moneyedness and remoteness. The more I think about it the more I suspect that the pilot market for this class of device would be the sick bay of a yacht that has its own Wikipedia page. Which is fine, but surely nothing that will have a meaningful impact on our general understanding of the epidemic.
Give people at-least some credit... :) It doesn't work like you think at all. We buy lots of high end equipment for our vaccine research/manufacturing, including PCR machines. Nobody in biotech opens up their wallet for high-ticket items based on a few words on a website. They have to actually demonstrate that it does what you want. We demand application notes, papers, data, industry references, the works. Before a purchase is made a URS document is prepared, then there is the FAT, the SAT, IQ/OQ/PQ validation and on and on..
I'm seeing a lot of grumbling here about this being "slow", but currently most COVID testing involves shipping samples around to remote labs. A NYT article about a hospital in Brooklyn mentioned that their samples are tested in California (!), and with these labs also inundated, average time to get a result is over a week now.
So if you can install these machines directly in hospitals, clinics etc, they will still be far faster than sending them off to a lab.
Practicing hospital physician here. We never use the term "accuracy" to describe diagnostic tests, and whenever I see that term used, I'm very suspicious of the claim being made. In other words, this Bosch announcement can probably be ignored.
We use the terms "sensitivity" and "specificity" to describe diagnostic test performance, and never to my knowledge talk about a test's "accuracy." Usually accuracy refers to the sensitivity, but there's a lot of liberty taken with that term in marketing talk.
For example, the existing PCR test being used does not have ideal sensitivity, which for practical purposes means that there is a substantial false-negative rate. Which is why (among other reasons) we weren't testing asymptomatic patients in the early phase.
According to Wikipedia, accuracy is actually defined in this context. This does not not mean that Bosch has not used everyday speech here but it is not clear that they have. Too bad that the article does not include links to additional information.
Of course accuracy is defined, but it's not the right metric because you usually have a large bias in the tested population: even if you only test people with symptoms, you might only have 10% of people that actually have the disease; so a test that always says "NO" would have 90% accuracy.
That's why you look at True Positive Rate (sensitivity) and True Negative Rate (specificity) to factor that bias out.
If TPR is 95%, that means that 95% of the people that are actually infected (10% of the pop in the previous ex.) will be detected as such.
IF TNR is 20%, that means you'll mistakenly detect 20% of people that don't have it as infected, which would be really bad.
In other fields, people use Precision and Recall. Precision = % of times the test is right when it predicts a positive. Recall = how many of the real positives were detected as such.
> IF TNR is 20%, that means you'll mistakenly detect 20% of people that don't have it as infected, which would be really bad.
Would it though? It's my understanding that most people who test positive are still being sent home, not forcefully hospitalized. And given that we don't actually have a widely used pharmaceutical intervention protocol, there's no side effects to worry about from false positives (eg of the opposite of this: if we were giving every confirmed case chloroquine or interleukin). The worst case outcome from a high false positive rate (low TNR) would seem to me that people would be extra good at self quarantining.
If you're a healthcare worker, the cost of a false-positive could be high.
For them and others one cost is that they'll feel themselves somewhat immune when they're not and may not take precautions. It will also confuse study of the disease if it starts to look like your "2nd" infection can be much worse than your first.
Hospitals in an epidemic are major spreaders of the infection. If false positive results lead to patients being put together with real carriers, they will soon be infected.
From a recent report from Italy:
> For example, we are learning that hospitals might be the main Covid-19 carriers, as they are rapidly populated by infected patients, facilitating transmission to uninfected patients. Patients are transported by our regional system, which also contributes to spreading the disease as its ambulances and personnel rapidly become vectors. Health workers are asymptomatic carriers or sick without surveillance; some might die, including young people, which increases the stress of those on the front line.
It could mean that some of the major causes of the large death toll in Italy are the lack of personal protection equipment, and, possibly, laxity in following protocol.
It can happen everywhere if the growth continues, simply because nothing can cope with fast exponential growth. At the moment, wherever it is uncontrolled, it's around 3 days to double. Sounds like small numbers, just 2 and 3? That gives however a thousandfold growth in 30 days: everybody working with computers should be very familiar with the equation 2^10 = 1024 (it's 2^(30/3) == 2^10 == 1024, 30 being the number of days for the projection, 3 the doubling time and 2 the doubling itself).
Nobody has thousandfold more hospital beds and doctors ready, even less a million times more, which is the two month's growth.
Discussing other factors without first admitting the major one is obviously biased.
> there just aren't enough people to sustain two months' growth
It can be simply calculated: it is expected that without measures the growth to 70% of population would be continuous (very approximately, the end phase wouldn't, we're estimating the limit). So the target is 6e9 people. If we assume that 4/5ths are the people who remain undetected by our current sampling, we want to know the growth between the current known infected and the target which is then 1.26e9 people. Currently known are 0.5 million infected. So the fastest end of growth phase would be just: 2520 times or around 2^11=2048 == just 11 times 3 day doubling time, or 33 days.
The growth will surely not be always exactly 3 days however, so it will be slower, but still not less dramatic, because the resources are many, many times smaller, in the poorer countries many tens of times smaller.
In short it can be very, very bad, and that will be much longer than just a month, just not the exact growth as now.
See the papers from Imperial College London for the exact shapes of the curves and the examples of their speed and growth.
That may be highly specific to this period, where italy outside hospitals is in lockdown. As we move forward, people will use the tests to decide whether to go back to work, which makes them more critical
The substantial false-negative rate is also one of the reasons why covid patients weren't released from quarantine after a first negative test, since a single result doesn't pass the risk threshold. Multiple negatives in sequence give the desired certainty of someone really being covid-free.
Thank you! I've been bringing up the sensitivity issue a lot recently, since it seems to be disregarded by many upset at the situation. I have a question for you if you don't mind: when so many people complain about lack of testing in the US, do you know what false-negative rates would exist for the tests we'd expect to see (like what they see available en mass in other countries)? And would you know what that level would need to be for the test to be useful?
No, accuracy is the same here it is for any binary classification. accuracy = (true positives + true negatives) / all samples. A largely useless measure. As a single score F1 score is much better, but sensitivity and specifity are specified separately for good reason. It makes napking math much easier, for one thing.
So would sensitivity be the percentage of infected people who test positive and specificity be the percentage of non-infected who test negative (i.e. sensitivity = (1 - false negative probability) and specificity = (1 - false positive probability))?
Yes. The intuitive way I try to remember it is to think that, in just plain English, "How sensitive is this thing?" means "How likely is it to detect whatever it's supposed to detect?" (e.g.: How likely is this security camera to detect movement?)
"Sensitivity" is an inaccurate term, since for non-native English speakers, it could be misinterpreted as a term to mean some kind of emotional response.
"Accuracy" is a bit more scientific- leaning and therefore easier to translate.
Remember, COVID-19 is a global phenomenon, its not just relegated to the Anglo-sphere. Bosch, being a German company, is probably a bit sensitive to the accuracy of translations that will be needed to market this thing effectively around the world .. ;)
> Remember, COVID-19 is a global phenomenon, its not just relegated to the Anglo-sphere. Bosch, being a German company, is probably a bit sensitive to the accuracy of translations that will be needed to market this thing effectively around the world .. ;)
I was only trying to help the parent find a way to remember which one is sensitivity and which one is specificity...
Yes, and you perfectly demonstrated the confusion possible by selecting the wrong terms. Sure, 'specificity' and 'sensitivity' are scientific terms - but for marketing material, can be confused with emotional responses. Thus, I believe Bosch chose 'accuracy', since this translates into other languages more effectively.
...marketing material? To the non-English world? I was literally just explaining a potentially helpful thought process to an English-speaking HN visitor dev who lists his(?) public key on his profile... not trying to produce marketing material for a French waiter or something. Are you sure you're not the one confused here rather than him?
I'm only trying to point out that Bosch may have specifically chosen the term 'accuracy' instead of 'specificity' or 'sensitivity', which don't translate easily outside of the Anglo-sphere, and in my opinion your thread is a valid example of why those two terms don't get used to market this device, which is what kicked this thread off in the first place ..
It is more likely that they were trying do deliberatly write a vague statement, their (native) German announcement uses the same "Genauigkeit", instead of the relevant terms "Sensitivität" and "Spezifizität".
Bosch sells its releases in 100's of different countries. I'm not saying they weren't using inaccurate terminology intentionally, more that there are indeed some guidelines for how things are couched in releases of this nature and one of those guidelines is translatability.
Especially apropos life-support systems where, indeed, RTFM||die is a thing.
I don't really know data science but in assay validation, sensitivity basically measures false negative rate, and specificity measures false positive rate. Accuracy and precision are measures of assay consistency or tightness. (statisticians may not purely agree with my simple explanation)
This article  suggests that false negative results will not be as problematic if we manage to start screening on a population level.
Inaccuracy of up to 15% could still keep the viral R0 under 1.0 (the critical value for eliminating an epidemic).
From the article:
> "False negative rates up to 15% could be tolerated if 80% comply with testing, and false positives can be almost arbitrarily high when a high fraction of the population is already effectively quarantined"
Shouldn't we optimize for a higher number of false positives? In the end, sending someone in quarantine for 2 weeks without them being infected is probably less dangerous than a false negative. So making tests overly sensitive would seem more applicable to me here.
"(...)the rapid test can detect a SARS-CoV-2 coronavirus infection in patients in under two and a half hours"
(...)With the tests currently in use, patients must usually wait one to two days for a result."
That approach makes sense for when you want a low false positive rate, i.e. when the treatment is dangerous or expensive (so you want to avoid it at all costs if it's unnecessary). Here, it's the other way around. Treatment is benign (stay at home unless it becomes an emergency), but a false negative is expensive (you'd infect lots of others). What cheap test do you know that can have a low false negative rate? The only one I know is the one that flags everyone as infected.
Given the size of hospitals around the world, I'd assume its not unlikely that they'll buy 100+ of these devices. I agee that this is somewhat confusing, but it doesn't seem too unrealistic.
Also, it seems like the actual test is the cartridge which is inserted into an existing device. So the device is probably already in use in some places and can get this extra test capability by just inserting the cartridge.
There's already a 45-minute rtPCR test from Cepheid and a simple 10 minute ELISA test from SureScreen (which serves a different need, antibody detection rather than infection detection).
It doesn't appear to have an FDA EUA so it may not be immediately usable in the US.
The user-friendly UX and compact size seem good.
The cycle time seems to be on the slow side when minutes count in an ER / ICU situation. ~9 tests per day per machine seems slow, costly and likely to occupy a lot of space for the machines and the cartridges (which contain reagents that have to be kept chilled).
This seems overall like a timely product to solve immediate needs. It could've been better but it was designed quickly.
>In various laboratory tests with SARS-CoV-2, the Bosch test delivered results with an accuracy of over 95 percent.
Curious what the false negative rate is.
>[The device] can perform up to ten tests in the space of 24 hours
>[Testing] can be performed directly at the point of care. This eliminates the need to transport samples, which takes up valuable time. ... With the tests currently in use, patients must usually wait one to two days for a result
>... even medical personnel who have not been specially trained on it can reliably perform the test.
>... it allows a single sample to be tested not just for COVID-19 but also for nine other respiratory diseases, including influenza A and B, simultaneously
This is fine, but it is yet another molecular test. What we need now are serology tests. They are faster, cheaper, easier to execute (blood-based), can detect cured patients, and can scale to the population of a country.
Serology tests are already used in several countries in Asia, I have little doubt they will arrive in Europe and the US soon-ish.
We already have serology tests in Czechia (got several days ago from China), but there is some confusion about how useful they are. They may produce negative results even for patients with several days of clinical symptoms, so negative results still needs to ve verified with molecular test.
”The widely available Roche’s cobas 6800/8800 Systems, which are used to perform the cobas SARS-CoV-2 Test, provide test results in three and half hours and offer improved operating efficiency, flexibility, and fastest time-to-results with the highest throughput providing up to 96 results in about three hours and a total of 384 results for the cobas 6800 System and 960 results for the cobas 8800 System in 8 hours. The test can be run simultaneously with other assays provided by Roche for use on the cobas 6800/8800 Systems.”
This is just another test cardridge available for the existing on-site lab machine. The idea is that health-practitioners with access to the device can now test on premise, instead of sending samples to the lab.
I can't image a lot local doctors have access to qPCR, where this device is useful for all kinds of other test.
I honestly don't even understand what I'm reading anymore. Either on HN and Reddit, or sites like the BBC, NY Times, and CNN.
Half the things I read still reflect fantasies of Asian-style containment in the West.
The other half acknowledge that this is too late. That we're heading up an exponential curve now, and will be on the other side in 3-6 months (depending on measures taken).
I'm no expert. But knowing what I know about my nation's government, and the likelihood of its people fully cooperating with any measures, I suspect that the second half is more realistic.
Meaning that testing and contact tracing is really only going to become relevant later this year. To snuff out any minor flare-ups that occur after most of society has gotten over the curve and built up herd immunity.
this whole every-day news about rapid test dev, etc is getting frustrating.
My SO is an MD in NY. All they want to hear from any company is, we have developed method that can be used to test 10k/day with results in 1 hr and will be mass producing and provding in two days.
That's already being done in some places that have limited numbers of tests. The governor of Nebraska announced that they are pooling tests to save on test kits. Samples from 5 people are combined in 1 test. If the sample tests negative, those people are in the clear. If the sample tests positive, all 5 are then tested again separately.
Making a fraudulent test now seems like a wonderful way to simultaneously piss off politicians (i.e. They'd be looking for sympathy points by reacting harshly) and literally decimating your market cap.