Wuhan Coronavirus Resource Thread

[SRoy]

It seems that's you here. I will let the PM know, he can rein you in.

Sure. Let ,me know.
Coming from the same people that promised, numbers peaking by Mid/End May (NITI aayog ??) 5T economy, 15L in bank accounts yada yada.

I wouldn't stoop to you level of cheap ad hominem barbs.

[Amber G.]

I saw the presentation. https://www.facebook.com/iitkanpur/vide ... 449404439/
They talked about 14% anti-bodies. If that is 17 crores and deaths are 100,000.

Based on that mortality rate will come down to 0.1% or less. They won't say 2%. Where did you get that?

Per their numbers .. ICMR survey (end of August) gives the value of 7% (These are the actual numbers from the survey - which obviously is under-count)
The actual number (parameter to fit the data - thus a reliable number) was around 14% around end of August.

At the peek (right around now) this is about 30%

(This is NOT the total "infected" people - this includes people have no symptoms and those who are not even aware that they have antibody - As Sudarshan pointed out "A" of ASIR model)
Mortality rate (projected) = LESS than (.04% of those who are known to be infected)

[sudarshan]

Coming from the same people that promised, numbers peaking by Mid/End May (NITI aayog ??) 5T economy, 15L in bank accounts yada yada.

Judging the merits of the current model by some list of mistakes (real/ perceived) made by its authors in the past, is silly. By that logic, we should deny every success of ISRO/ DRDO because they made so many mistakes in the past.

If you have any real issues, please tell (not just the fact that the model shows disagreement with numbers you pulled out of someplace).

As a matter of fact, there probably was a peak end May/ early June, like I keep saying. Just that there was another peak on top of that, so there was a distinct inflection point.

Even if that was not the case - it's very hard to predict with a pandemic, especially with a totally unknown virus, so your impossible standards of past accuracy are not the metric to judge the current model here.

I wouldn't stoop to you level of cheap ad hominem barbs.

* You started your previous post by suggesting that a poster here (Amber G) was in a state of intoxication
* You referred to the authors of the model as "clowns" who needed to be "reined in"

When that last statement is turned back on you, that becomes "ad hominem."

[disha]

It seems that's you here. I will let the PM know, he can rein you in.

Sure. Let ,me know.
Coming from the same people that promised, numbers peaking by Mid/End May (NITI aayog ??) 5T economy, 15L in bank accounts yada yada.

I wouldn't stoop to you level of cheap ad hominem barbs.

SRoy'ji, do you have another model? What does your model project? Also can you put your model out?

As an outside observer, the team AmberG refers to at least have put their necks out by creating the SAIR model and providing projections. They seem pretty much on track.

I have also been putting out some charts for NY and CA. And I have not seen a model that given the data correlates it with past data and tries to project onto future.

Hence the model for India is indeed remarkable. Unless you have a competing model which you are going to present, all your statements can be construed as political motivated.

[Vayutuvan]

Press conf link <From IITK Site .. Hope this works for public. (I am sure there will be a youtube link shortly ..if so I will post it)

Extremely clear cut and lucid explanation (as usual) from Prof. M. Vidyasagar sir. What we had in the US was the Gov. Cuomo doing PR as well as Whitehouse doing PR.

I hope somebody who is a good teacher from MIT or some other place gives a news conference of this sort in the US.

[SriKumar]

It seems that's you here. I will let the PM know, he can rein you in.

Sure. Let ,me know.
Coming from the same people that promised, numbers peaking by Mid/End May (NITI aayog ??) 5T economy, 15L in bank accounts yada yada.
I wouldn't stoop to you level of cheap ad hominem barbs.

I am not sure but I doubt this team in IIT-K (or whoever built this model) came up with the predictions of peak in May, AND 5T economoy AND 15L in bank accounts. Since you lump them all together, I suppose 'same people' refers to PM Modi, or Home Minister Amit Shah - 'clowns to be reined in'. Odd that you are sensitive to 'ad Hominem' comments.

Anyone one who puts out a well-reasoned, vetted and verified mathematical model for a national crises that points to a future direction, either a positive one or a negative one, is doing a service. FOr me, this model is important because it tells us that (i) if the ppopulation continues to use masks and does a,b,c, there is a good chance the epidemic will end by February. This also gives people a direction on what they can do (durnig the festival season) to keep things under control. If the actual disease numbers buck the prediction in Novermber, it allows the govt to remind people or increase inforcement/quarantines to keep things under control. If the actual disease numbers follow the curve, the govt needs to do nothing, and people are reassured that the end is in sight (in Feb.). This model gives us all a way to track if and when things start to get out of control so govt can intervence rapidly- very important in a national crisis that involves every single child, woman and man.

At a minimum, authors of the model (not PM or HM) have made a public prediction, and if things go seriously off prediction, the professors/mathematicians/anyone with their name attached to the work have staked their professional reputation in a very public and honest manner.

[Amber G.]

[

Extremely clear cut and lucid explanation (as usual) from Prof. M. Vidyasagar sir. What we had in the US was the Gov. Cuomo doing PR as well as Whitehouse doing PR.

I hope somebody who is a good teacher from MIT or some other place gives a news conference of this sort in the US.

Prof M. Vidyasagar is world renowned and respected professor/scientist/mathematician (FRS, distinguished professor at IIT etc).. He is son of famous Canadian/Indian mathematician/number theorist M. V. Subbarao.

Manindra Agrawal ( Infosys Prize, Shanti Swarup Bhatnagar Award, Padma Shri) is of AKS primality test fame - (I have talked about him quite a few times in math dhaga).


Modi/GoI, selected them as leaders and they were able to put a great team consisting of doctors/Engineers/economists etc. I think they were appointed in late may (or early June) and Herculean amount of work (incorporating/condensing about a dozen models used in the world around that time) in a very short time. The result were published in late June.. their work was finished but GoI/DST was quite happy and tasked them to do further continue their work - advising on Medical inventory management, and other modeling.

All this work was quite transparent - findings regularly published etc..

[sudarshan]

...

Responded in "Psy-ops techniques" thread, since it is OT here.

[SriKumar]

^^^ Read it.
Am curious about your take on WHO’s statement on Remdesivir. I have not been following this thread for a month or two and hence missed your reasoning/hints.

[sudarshan]

^^^ Read it.
Am curious about your take on WHO’s statement on Remdesivir. I have not been following this thread for a month or two and hence missed your reasoning/hints.

I didn't actually present my reasoning so far in this thread, but here it is, FWIW.

I don't think of the WHO as being focused on "World Health," their initial duplicitous role in this pandemic was pretty evident (letting China off the hook and not imposing a travel ban, among other things). I still don't think their motivation is the well-being of citizens of the world.

That being the case, dissing HCQ was the obvious thing to do (no money in it, it's what - $5 a dose?). Now look at the models for India that Amber G posted here. There must be similar projections going on for the rest of the world, there hasn't really been a second wave in Europe (case counts notwithstanding - I'm talking of deaths, serious illnesses, hospitalizations). So the projections the world over probably show a collapse of the market for COVID treatment in general, and remdesivir in particular (at least WRT COVID), beyond the current year.

COVID treatment goes by actual demand, people look for it when they really need it. Vaccines on the other hand go by fear psychosis and perceived demand. If the projections show that remdesivir would be needed in maybe millions of doses for the next year for COVID, but vaccines - billions of doses could be sold per year (with three or four booster shots per year) - over a period of five or ten years - which market would be expected to be larger? That's one reason why I was asking Amber G about the perceived need for vaccines in India beyond Feb. of next year.

So, build up the perception that there is no cure or treatment, only vaccines will help. Now go look at that headline about the 89-year-old Danish (or Dutch?) woman (cancer patient) who died of a COVID reinfection, and see how the doctors in that article extrapolate that reinfection risk to the general population. Also look how they take the four or so documented incidences of reinfection the world over and build on it.

I was seeing the long-term trend of collapsing death counts from the data, and I thought probably the WHO was seeing the same thing and hedging their bets accordingly. That's why I was encouraged that the "official" projections at the WHO were also showing a collapse of deaths/ hospitalizations/ severe disease incidence with COVID.

Maybe just my paranoia, but you can see for yourself, the way the WHO has been behaving all through this pandemic.

[Amber G.]

^^^Similar thoughts (about WHO) have been expressed by others who have studied the data and IMO knows much more than I.

(Prof Vidyasagar, for example, doubts very much about the statistical implication of the data about reinfection - some cases (dozens out of millions, btw) they are not even sure, if the case is actual reinfection (rather than continuation of the original infection) - please don't quote me as I may be wrong about what he said/thinks exactly - but read the actual analysis from them)

Sudarshan, you may already know, but the authors (and most of the scientific community under present admin) are quite approachable and have shared their findings in various platforms widely. For example FAQ section of link I gave before (covid19medinventory.in) has email address of the authors and good writeup about how to use the tools using any data (and even your own model) to analyze it etc. Also IIT and IISc group are also are quite approachable.

FWIW, I do hope WHO may get reformed with new coming leadership (Dr Harshvardhan).

[hnair]

sudarshan, brace yourself for a version 2.0 of “HCQ the demon medicine” campaign by big pharma, now that the GoI funded mathematical modeling findings say things will taper off by Feb!

I mean, if we are talking of inoculation of only 65+ age group and health workers, then a few billions of dollars in stocking up of retro-virals, testing kits and even the vaccine shots have vanished into thin air. I fully expect scare mongering around Deepavali super-clusters, “PM Modi believes in pseudo-science offered by right wing professors” etc by pharma/media

We saw right before our eyes, the politicization of science by a political side that claims to be for pro-science, under encouragement from Big Pharma to I sell anti-viral by the billions over a 5cent pill!

Two essays by same author on the HCQ sagas:
https://www.tabletmag.com/sections/scie ... sts-doidge

https://www.tabletmag.com/sections/scie ... ality-tale

(Got this as a foreward from a prominent columnist)

[vijayk]

We saw right before our eyes, the politicization of science by a political side that claims to be for pro-science, under encouragement from Big Pharma to I sell anti-viral by the billions over a 5cent pill!

Two essays by same author on the HCQ sagas:
https://www.tabletmag.com/sections/scie ... sts-doidge

https://www.tabletmag.com/sections/scie ... ality-tale

(Got this as a foreward from a prominent columnist)

Very objective

Very sad ... The vicious propaganda of lies by US media against HCQ is one criminal act. I just want Americans to teach a lesson to Dems just for this reason. The propaganda war by scum media and their total connivance with Chinese CCP to absolve them of all the crimes against humanity has to be punished by people. Alas! Might not happen.

[hnair]

No US politics please. Big Pharma has no sides - they will align with all sides in this

[Amber G.]

PM needs to rein in these clowns.

LOL! I think PM trusts and respects these "clowns"..Virtually the same bunch of "clowns" ( For example - Mahindra Agrawal and others - also led/inspired/guided the other teams) produced these "Swasa" masks which you see Modi wearing while doing bhomi-pooja at Ayodya. (See Times of India story : PM Narendra Modi wears IIT-Kanpur’s Swasa N-95 mask



(Not only Modi, Kangana Ranaut and CMs of the many Indian states (including non BJP) too trusts these masks
Also, doctors, our armed forces, police etc..-)

(Note: I am *not* taking away credit from *many* others (outside IIT/IISc etc) who worked *very* hard to fight this national pandemic - There are, obviously *many* heroes here - personally I have never seen such good team work in India as I have seen in these last many months. I really hope this new culture (of doing good work for general benefit, rather than getting credit on a research paper) stays with Indian scientists)
***
Seriously the story is very inspiring : How this team - a professor/engineer and some of his students - started the idea - bunch of alums poured the money and expertise, Best schools and scientists in US helped, - GoI helped, CSIR approved many items without usual red-tape, Doctors, Hospitals and Big Industry helped and within months, India which did not produce *any* N95 Masks or PPE in India, and have to import all this in March etc.. now has a capacity to produce few hundred thousand masks and PPE kits per day - and produces, from last I heard, few thousand masks and PPE's per day.

Of course, this is not an isolated story - these same group of dedicated Indian scientists have similar story for Ventilators, testing kits, research in Vaccines etc..

[Amber G.]

Some may be interested in this very interesting conversation at 4.30PM (India Time) on 21st Oct, 2020.

[suryag]

AmberG garu how is Prof.Abhay doing, trust he has recovered well

[Ambar]

Sudarshanji, while i have zero faith on WHO and absolutely don't trust them to do the right thing, i dont think there's a grand conspiracy to discard HCQ/Remdesivir etc. Despite the times we live in where healthcare is big business, there are many doctors everywhere who take their pledge to their profession seriously, and they give two hoots about what WHO (or for that matter ICMR / MoHfw) has to say. They will administer what is necessary if it saves lives. HCQ has been tried but there is no conclusive evidence that it works, and on the other hand i've been unfortunate enough to witness first hand the empirical evidence of the adverse effect of HCQ. If it was HCQ for us, it is Ivermectin in Latin American countries and there too we've seen little to no evidence that it actually works. Remdesivir continues to be the first weapon against covid, but its efficacy largely depends on the age, comorbidities, condition of the patient and timeliness of administering the drug. It does little to stem the mortality of severely ill hospitalized patients, where as it does seem to slow the virus down if given between day 2 and 5 before the lung gets infected and turns into pneumonia. Until recently due to scarcity of Remdesivir in India doctors were not administering it unless the patient was seriously ill, but recently i've seen friends and relatives (two of whom are currently in hospital as a precaution) were given Remdesivir right from day 1 after their covid test turned positive.

WHO is not entirely wrong by saying there is more hope on vaccines slowing the pandemic than finding a sureshot cure. No antiviral or immuno suppresive/anti-parasite drugs have proved effective against Covid, where as a vaccine even if it works only for 50%-60% of the population will still prove to be more efficient in containing the virus.

The 2nd wave in Europe is real, deaths alone should not be the parameter because doctors & hospitals have learnt much about the virus in the last 6 to 7 months ,hence most deaths even in US occurred in Mar-Apr period. Secondly, after seeing the devastation many elderly people and those with comorbidities are now careful, so the new cases in Europe and US are in the relatively healthier young population which also explains less hospitalizations/deaths. Having said that one only needs to look at the rising cases, hospitalizations and deaths in UK, Spain, France and last few days in Italy to see that a 2nd wave is indeed sweeping across Europe, this is also the reason why Modi came on TV to warn against complacency in India.

I think it is fair to say that the worst may be behind us in terms of deaths but for countries like India if & when there is a 2nd wave and if the number of cases are higher than the peak we saw in September then even a 1% CFR may be devastating. I am also very surprised that Pakistanis and Bangladeshis did not drop like flies like i thought they would due to the complete negligence, incompetence and general baboonery of its populace and their governments, maybe those 2 countries hold the key as to why richer Indian states seem to have suffered so much more than the poorer & poorly run ones.

[Cyrano]

We saw right before our eyes, the politicization of science by a political side that claims to be for pro-science, under encouragement from Big Pharma to I sell anti-viral by the billions over a 5cent pill!

Two essays by same author on the HCQ sagas:
https://www.tabletmag.com/sections/scie ... sts-doidge

https://www.tabletmag.com/sections/scie ... ality-tale

(Got this as a foreward from a prominent columnist)

Thanks for posting this hair ji. Incredible times we are living in. This pandemic is bringing out the best and the worst of people in positions of responsibility and power.

[Amber G.]

^^ @suryag -- Thanks. He has recovered (it was fairly serious and we were all very worried) and doing great. He is back to being extremely busy and had been in spot light. Just in last week or so, he was induced in TSDSI Hall of Fame as a "TSDSI Fellow" and IITK honored him with a special award. IBM's CEO (an IITK alum) will be giving the convocation address at IITK .. and among other things from the recent news item - he is doing a bhoomi-pooja of new research park- Technopark IITK. With a vision of co-developing indigenous cutting-edge technologies in partnership with industry and further bolstering a robust research and technology development ecosystem in the country..(I am glad that after months of Covid slowdown the construction/expansion work is starting ..I do believe these guys do believe in Mathematical modeling and science to start all this work)

[sudarshan]

The 2nd wave in Europe is real, deaths alone should not be the parameter because doctors & hospitals have learnt much about the virus in the last 6 to 7 months ,hence most deaths even in US occurred in Mar-Apr period. Secondly, after seeing the devastation many elderly people and those with comorbidities are now careful, so the new cases in Europe and US are in the relatively healthier young population which also explains less hospitalizations/deaths. Having said that one only needs to look at the rising cases, hospitalizations and deaths in UK, Spain, France and last few days in Italy to see that a 2nd wave is indeed sweeping across Europe, this is also the reason why Modi came on TV to warn against complacency in India.

Read the entire post, responding piece-meal as and when I digest.

I looked up the testing numbers in Spain, France, etc., and you're right, the wave there seems to be real, since the positivity fraction has risen sharply. That's definite cause for concern.

In India, case counts are dropping, but not because of testing drops. The positivity fraction is dipping, almost in proportion to case counts (since testing is relatively flat). So the drop is "real" in that sense.

This is a nice website, where they color-code the positivity fraction onto the case count curves for various countries, so you can instantly see not just case count rises, but their relation to testing as well. (Scroll down for the: Daily confirmed cases: are we bending the curve? plot).

https://ourworldindata.org/coronavirus/ ... untry=~ESP

EDIT: OK, which one of you is going to change your handle? Going nuts keeping things straight.

@Ambar, @Amber G

And on top of that, there's this @amar_p feller as well.

[Amber G.]

^^^Thanks ... Hi this is @amber G. - the most ancient - have been in brf since 1998.. . (Mostly visible in Math/Physics/Nuclear/Science related dhaga's )

FWIW - in Math dhaga I posted "how to use SIR model. For those who may not know, it explains briefly what is SIR model and how one can use it. There are many , even free, tools where you can analyze/project etc..

India/DST's "super" SAIR Model, as said here before has one more parameter: Here are the basics - Mathematical details are in the paper I referenced in earlier post.



You can run this against any data - using your own values - and see the projection (or verify how accurate you numbers are by looking at old data and how it fits)

(If solving differential equation is not your cup of chai - many mathematical programs/tools etc can be used to produce those beautiful graphs)

[vijayk]

This telugu newspaper says artesunate has been very effective in curing COVID-19 but Govt. of India and the world ignored this. One doctor in Vizag has used to cure all cases.
Saudis are using it very much but GOI is not recognizing this

Any one heard about it?

https://clinicaltrials.gov/ct2/show/NCT04387240
Evaluating the Efficacy of Artesunate in Adults With Mild Symptoms of COVID-19

https://www.cell.com/trends/parasitolog ... 20)30288-9
Repurposing antimalarials to tackle the COVID-19 pandemic

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7254722/
Artesunate: could be an alternative drug to chloroquine in COVID-19 treatment?


https://medicalxpress.com/news/2020-07- ... risky.html
Artemisinin: fighting coronavirus with this antimalarial drug is risky

https://www.sciencedirect.com/science/a ... 1220306615
Antimalarial artemisinin-based combination therapies (ACT) and COVID-19 in Africa: In vitro inhibition of SARS-CoV-2 replication by mefloquine-artesunat

[Amber G.]

This telugu newspaper says artesunate has been very effective in curing COVID-19 but Govt. of India and the world ignored this.

To be fair, about GoI - there have been 36,000+ research papers on Covid related stuff which is being looked at by GoI scientist ..it is quite a task to sift through all that -- much harder than writing an article and criticizing/accusing others of ignoring them. JMHO

[Amber G.]

Some may be interested in this very interesting conversation at 4.30PM (India Time) on 21st Oct, 2020.<info "Other Side of Pandemic">

Quite Nice - really liked it. Panel consists of top scientists/economists/Medical and communication experts. Worth watching.

Here is direct youtube link: https://youtu.be/pIKdEcsvdWw

[sudarshan]

Amber G, appreciate the parichay . I responded to a post by Ambar (I thought) and that turned out to be you. Then my previous post was a response to you (I thought), some time later when I looked at the post, it said "Ambar wrote" at the top - that's what the outburst was about. As for amar_p - that's a name which the admin gave him, because the admin thought his initial choice of name referred to a no-SQL database....

Anyways, about the SAIR model, some comments:

The model needs an initial state, right? That video in the math dhaga covers that in better detail (where the guy sets a variable to be 1, which represents the full population and also assumes initial values for S, I, R).

Why not just difference the model, why solve the differential equation (analytically, I think is what you were suggesting)? It could be differenced with a time step of 1 day, since that's the reporting period for all countries. Then it would be kind of like a Markov model, but not really, since there are some terms with products of different states (like SI*(A+I)). This would be a lot easier for somebody who wasn't that familiar with diff. eqns, or who was intimidated by them.

The parameters (beta, gamma, etc.) reported in the Supermodel site don't seem to have units, I'm guessing those are on "per day" bases.

The parameter epsilon seems pretty unrealistic. They report 1/eps=1000 for phase 1. That means (roughly) that for every infected (and symptomatic) person, there would be 1000 asymptomatic persons? Is the ratio that high? Even for phase 3, 1/eps was reported to be like 64, which means (roughly) 98.5% of infections are asymptomatic, and only about 1.5% actually develop symptoms? Or is epsilon only required for the initial state setup (I saw that it referred to SI_0 and SA_0)?

When you said - "anybody can try the model out with their own data" I got a very wrong impression. I thought anybody could take their own timeseries data and feed it to some engine on the website, and the engine would infer the model parameters (maybe by least-squares fitting) and fit the entire data to that. That would be something, and would be much more lay-person friendly. But it would require some intelligence on the part of the engine to figure out different phases (inflection points) and fit different parameter sets for each phase. Whereas, I think what is being talked about, is that anybody can assume their own values of the parameters, and use the equations to derive timeseries output, which can be compared with actual data. That's not so lay-person friendly, but more mathematically-inclined folks would like that.

In any case, having the model out in the open should help to some extent against propaganda like hnair saar was saying - "Hindoo nationalist Modi employs right-wing 'scientists' to come up with pseudo-scientific predictions." Not that hard-boiled cases like the BBC would let that stop them, but some less loony outfits - maybe.

[nandakumar]

Amber G, appreciate the parichay . I responded to a post by Ambar (I thought) and that turned out to be you. Then my previous post was a response to you (I thought), some time later when I looked at the post, it said "Ambar wrote" at the top - that's what the outburst was about. As for amar_p - that's a name which the admin gave him, because the admin thought his initial choice of name referred to a no-SQL database....

Anyways, about the SAIR model, some comments:

The model needs an initial state, right? That video in the math dhaga covers that in better detail (where the guy sets a variable to be 1, which represents the full population and also assumes initial values for S, I, R).

Why not just difference the model, why solve the differential equation (analytically, I think is what you were suggesting)? It could be differenced with a time step of 1 day, since that's the reporting period for all countries. Then it would be kind of like a Markov model, but not really, since there are some terms with products of different states (like SI*(A+I)). This would be a lot easier for somebody who wasn't that familiar with diff. eqns, or who was intimidated by them.

The parameters (beta, gamma, etc.) reported in the Supermodel site don't seem to have units, I'm guessing those are on "per day" bases.

The parameter epsilon seems pretty unrealistic. They report 1/eps=1000 for phase 1. That means (roughly) that for every infected (and symptomatic) person, there would be 1000 asymptomatic persons? Is the ratio that high? Even for phase 3, 1/eps was reported to be like 64, which means (roughly) 98.5% of infections are asymptomatic, and only about 1.5% actually develop symptoms? Or is epsilon only required for the initial state setup (I saw that it referred to SI_0 and SA_0)?

When you said - "anybody can try the model out with their own data" I got a very wrong impression. I thought anybody could take their own timeseries data and feed it to some engine on the website, and the engine would infer the model parameters (maybe by least-squares fitting) and fit the entire data to that. That would be something, and would be much more lay-person friendly. But it would require some intelligence on the part of the engine to figure out different phases (inflection points) and fit different parameter sets for each phase. Whereas, I think what is being talked about, is that anybody can assume their own values of the parameters, and use the equations to derive timeseries output, which can be compared with actual data. That's not so lay-person friendly, but more mathematically-inclined folks would like that.

In any case, having the model out in the open should help to some extent against propaganda like hnair saar was saying - "Hindoo nationalist Modi employs right-wing 'scientists' to come up with pseudo-scientific predictions." Not that hard-boiled cases like the BBC would let that stop them, but some less loony outfits - maybe.

Off track. But couldn't resist the temptation to make a wise crack. What you need is a spell as sub-editor at one of the mainstream print media to spot subtle differences in the written word!!?

[sudarshan]

^ Nukkad-76 thread.

Amber G ji, if the discussion above is too mathematical, we can continue in the math dhaga.

[Ashokk]

Quick Covid-19 test from IIT Kharagpur: How it works, why it matters

A new diagnostic test for Covid-19 from IIT Kharagpur, called COVIRAP, can be a potential game-changer in bringing high-end molecular diagnostics from the lab to the field. The Indian Council of Medical Research (ICMR) has validated the efficacy of the method, IIT-KGP said Wednesday.

How it works: It has an automated pre-programmable temperature control unit, a special detection unit on genomic analysis, and a customised smartphone app for results.

Three master mixes work as markers of different genes to confirm the presence of SARS-CoV-2, said Prof Suman Chakraborty, Dean, Sponsored Research and Industrial Consultancy, IIT Kharagpur. Samples collected react with these mixes. When paper strips are dipped into these reaction products, coloured lines indicate the presence of the virus.

Dr Arindam Mondal, Assistant Professor, School of Bioscience, IIT-KGP, said the technology has been subjected to rigorous testing protocols as per ICMR guidelines.

Why it’s special: Current tests include RT-PCR, which are highly accurate but require advanced lab infrastructure, and antigen tests that can give results in minutes but have a lower accuracy.

Prof VK Tewari, Director, IIT Kharagpur, said the COVIRAP process is completed within an hour. The test is conducted in a ultra-low-cost portable unit that can be handled by unskilled operators outside the lab environment and is an alternative to high-end RTPCR machines. It can test samples even on open fields. The same unit can be used for a large number of tests on replacement of the paper cartridge after each test.

The patented machine unit is also very generic, which means that it can perform tests beyond Covid-19 — for influenza, malaria, dengue, Japanese encephalitis, TB etc, under the category of isothermal nucleic acid-based tests.

COVIRAP & FELUDA: FELUDA, named after Satyajit Ray’s fictional detective as an acronym for FNCAS9 Editor-Limited Uniform Detection Assay, is a test developed by the Institute of Genomics and Integrative Biology. This too detects genes specific to SARS-CoV-2, but uses CRISPR-CAS technology.

With FELUDA, too, the need for technical expertise is minimal. While the current FELUDA prototype requires a PCR machine for processing, COVIRAP uses its own detection technology, patented by IIT-KGP. “There are certain components of the method which are exclusive for us and different from CRISPRCAS,” Prof Chakraborty said.

[Cyrano]

Sudarshan wrote:

As for amar_p - that's a name which the admin gave him, because the admin thought his initial choice of name referred to a no-SQL database....

LoL !! Do admins change handles once assigned ? I'm OK to change it to something without the DB part !

[Rahul M]

no issues. just select a human sounding name & let us know.

[Cyrano]

Thank you Rahul ji, could you please change it to Cyrano ?

[Amber G.]

Quick Covid-19 test from IIT Kharagpur: How it works, why it matters
.

Thanks. Interesting tidbit - "FELUDA" (Detective from from Ray's movie) name was given by a wife (from Maharastra) of a Professors from (Bengal.)...

[chola]

I looked up the testing numbers in Spain, France, etc., and you're right, the wave there seems to be real, since the positivity fraction has risen sharply. That's definite cause for concern.

In India, case counts are dropping, but not because of testing drops. The positivity fraction is dipping, almost in proportion to case counts (since testing is relatively flat). So the drop is "real" in that sense.

This is a nice website, where they color-code the positivity fraction onto the case count curves for various countries, so you can instantly see not just case count rises, but their relation to testing as well. (Scroll down for the: Daily confirmed cases: are we bending the curve? plot).

https://ourworldindata.org/coronavirus/country/spain?country=~ESP

Sudarshan ji, can we safely say we have peaked and are on the other side of the mountain?



What's more, this is happening in the midst of Unlock 4.0. Cases are going down even as we open up!

I do not want to sound overly confident but India is beating this thing while the US and Europe are losing grip again.

[Cyrano]

While the stats are indeed encouraging, I have my doubts about how precise, accurate and complete the ground up data collection is, in India and in many countries in the world. All models are only as good as the reliability of the base data. One could argue that this is a persistent noise factor and that as long as it stays relatively constant the resulting trend would still be relevant and there is indeed some truth in that. In some matters it helps, in some others like gauging preparedness for the next wave, it could throw us off.

One way of seeing whats happening in the US is to view it as an accelerated script of whats unfolding in the rest of the world. There will be multiple waves of infection spread, the size of each successive wave depending not only on the measures taken/imposed and improvement of treatment protocols, but also inevitable social factors like summer holidays, schools & colleges reopening, festivals in India like Dasara and Deepavali, X-mas and BlackFriday sales elsewhere, compounded by (trans-border) movement of people as countries will feel compelled to loosen restrictions as their economic resilience starts to run out. The net effect will likely be that each new wave being a bit stronger than what we can extrapolate from mathematical models, but overall (hopefully) weaker than the previous wave and peter out as it runs out of new persons to infect. Kind of like a ball bouncing on an uneven surface.

This could take few more quarters, may be a year or two from now until new infections become insignificant.

[chola]

^^^ Understood. Data might not be entirely accurate but trendlines taken over time usually are. At least that's how things work in the business world. I'm not a health expert or scientist but statistics tend to work the same way.

I would say the trendlines are exceptionally good. We are on the othe side of the slope with infections going downhill after mounting what looks like a definite peak in September.

Further more, there are no real negative mitigating factors to the drop -- lower testing and lockdown. If the lower numbers are because we are testing less or because we are going into progressively restrictive lockdowns then I would be more hesitant. A spike in testing or loosening of lockdown could bring numbers up again.

But testing has not dropped and we are not in lockdown but coming out of one for weeks now! And the daily infections are dropping even as the country opens up. So things are very encouraging.

[Cyrano]

Further more, there are no real negative mitigating factors to the drop -- lower testing and lockdown.

That is indeed a good thing.

[kvraghav]

Data should always be extrapolated unless there is a change in the way data is collected between the increasing phase and decreasing phase. The error in data collection would even be there in the ascending phase and we should only concentrate on the pattern.

[Ambar]

It has been posted before but India's peak flu season is between Apr and Sept, where as US is entering its fall and winter flu season now. While i have serious doubts on data collection especially in central and western states in India, there is no doubt that the cases are falling. I track several districts not just through ICMR's site but through local newspapers that are much closer to the ground, and they all have been reporting fall in hospitalizations and people in ICU. On the other hand US and Europe are clearly in a 2nd wave, the number of new cases in some Europeans countries is 3x to 4x times their peak cases in March and April ! Yet the deaths are nowhere near record highs possibly because the new cases are mostly in younger patients, and the anti-viral, anti-inflammation cocktails being administered now is reducing the fatalities.

In other news after WHO claimed Remdesivir does not work, FDA went ahead and approved it as the first treatment for Covid-19.

[sudarshan]

Sudarshan ji, can we safely say we have peaked and are on the other side of the mountain?

...Image deleted...

What's more, this is happening in the midst of Unlock 4.0. Cases are going down even as we open up!

I do not want to sound overly confident but India is beating this thing while the US and Europe are losing grip again.

We're definitely on the other side, my personal preference is not to be so cautious about admitting such good news to ourselves, even when it could be just temporary good news. The question is - is there another mountain coming up. As you point out, it's most encouraging that there isn't a general lockdown going on, and still there is this declining trend. I'm tempted to say "herd immunity." The heartening thing is that most of the NE US has been flat for months (positivity rate), even as some neighboring states are going up (like PA). Again, it could be temporary good news, but that is what has been going on - I don't see the point of fearing "jinxes." So it could happen in India also, that once this peak is fully over, there won't be any further resurgence.

Positivity rate in India is under 6% now, and still dropping. The WHO criterion is that any area with less than 5% positivity rate for a period longer than 2 weeks, can be considered as "safe" in the sense of "no need for lockdowns currently".

As others have pointed out, it is the trend which is important, so long as the noise pattern is consistent. AFAIK, India hasn't changed its reporting standards. So for now - the news from India is pretty good.

One of the most significant things I noticed, was that in India, deaths started decreasing almost at the same time as case counts. With testing flat - that leads me to conclude that there is no significant lag between cases and deaths, certainly not longer than a week or two. So it seems to me that these fears of "deaths are low right now, but since cases are going up, deaths could start rising a month or two from now and start overwhelming hospitals" - those are way overblown. I too feared that India was going to see death counts exceeding 3, or 4,000 per day (if one assumed a month's lag between cases and deaths) or even 8, or 9,000 per day (if the lag was two months or more), that simply didn't happen.

Current case counts are a reasonably good indicator of what is happening or what is going to come (so long as testing is consistent - which means that what one is tracking, is actually the positivity fraction, not raw case counts). OTOH, it also means that one can directly look at death counts (without fearing that there is something building in there which has already happened, which is currently still hidden, but which will cause those counts to go berserk a month from now), those also seem to be a reasonably good indicator.