Jordan Micah Bennett, software engineer/creator of "RobotizeJa".
The aim is to develop a quick way to detect the nCov 2019 (Coronavirus 2019/2020, also called "Covid-19") strain, with the plan to use artificial neural networks or other machine learning model types.
A viable image based path reasonably resembles: SMART-CT-SCAN_BASED-COVID19_VIRUS_DETECTOR
A viable genome based path could resemble:
-
(a) Dna from person to be screened → (b) Genome data from MinION device → (c) Trained algorithm that has been built to distinguish between nCov -positive genome data, and healthy or rather nCov-negative genome data → (d) prediction-classes: nCov[~1] or no-nCov [~0]
-
Based on the available data/nCov genome information, this solution may generate an optimal way of quickly identifying new nCov cases, and replace insufficient temperature/thermometer "screening" processes.
As this is the first known attempt, commencing on January 29 2020 aimed collaborating to construct this type of program, please point to open source packages with similar goals. Please email jordanmicahbennett@gmail.com
- This can also reasonably allow for less experienced medical personnel to make preliminary diagnoses, expanding the diagnosis efforts overall. This effort may contribute towards virus-control progress, together with other ai based endeavours being developed accross the globe, such as use of ai for vaccine development.
Coronavirus: Whole world 'must take action', warns WHO
Update Jan 31, 2020/WHO declares the new coronavirus outbreak a Public Health Emergency of International Concern
- WHO's warning should reasonably have come about a week earlier, as advised about a week ago via Chris Martenson, who I also refer to below regarding his 115 million nCov case prediction count.
- Update February 7, 2020: Artificial Intelligence Prediction: In 45 days, ~2.5 billion to be infected, ~52 million of which may die.. See also this detailed forbes report.
-
Take as input, human genome data. (Obtained in the form of a blood sample from target human being screened for nCov)
-
Train an algorithm to distinquish between nCov positives and negatives, using an artificial neural network together with the labelled genome information, taking labelled sets of nCov positive genome data, together with nCov negative/absent genome data.
-
Invoke trained algorithm with the ability to within milliseconds, return prediction or classification of new unlabelled genome data (i.e. a person at the airport or quarantine room), aka detect new cases with good accuracy/confidence, of the latest nCov/coronavirus.
- Note: Suggestions for other paths are welcome.
- By extension, apart from human radiologist detection, perhaps an ai based image detection solution can speed up diagnosis, and help to replace the faulty dna based comparison phase. I've also requested more CT image data from a scientist involved with manual diagnosis using CT scan data.
- The nCov 2019 (Coronavirus Strain 2019/2020) is spreading rapidly, with a mortality rate between 2% and 4%.
- By comparison, the common flu with a far lower mortality rate of .1%, kills 291,000 to 646,000 per year.
- Things get worse; nCov spreads at ~tripple the transmission rate of the common flu.
- Common flu RO = 1.28 (Estimated, transmission rate)
- nCov RO = 2.5 to 3.8 (Estimated transmission rate)
- Recent nCov RO estimate ~4.08!
- Recent nCov 2019/Covid19 incubation period is estimated at 24 days, and a Chinese woman was recently struck down with symptoms after probation period of 15 days, according to the sun newspaper!
- Current screening methods may miss the presence of the virus because:
- The incubation period may constitute 0 symptoms, so temperature scans may miss carriers with 0 symptoms during incubation.
- Over the counter pills can be used to lower temperature, again, averting the temperature scanning/screening measures.
- January 31, 2020 Update: As an example, there has been one confirmed Asymptomatic driven case, where the virus has been spread by a person without symptoms:
- This ai driven method will reasonably help to stop the exponential growth of the nCov strain.
- 1 more month of exponential nCov growth = ~ 115 million cases, (of which ~ 23 million are potentially life threatening ones) according to an epidemiologist/PhD pathologist.
Although I am now experimenting with Janggu and ViralMiner...(see associated paper), I am yet to entirely identify all the tools that may required for this project, and I have not yet determined if the steps are robust enough. Software developers/machine learning practitioners/Ai developers could join in to contribute.
I encourage, or rather, the numbers above indicate that a non-trivial percentage of programming time/effort be placed into this endeavour, via all volunteers.
- Using ViralMiner's already pretrained models wrt genome deep learning, may be worth the resulting efficiency gained in genome analysis and identification in human samples, related to nCov coronavirus 2019 genome datasets released via China seen below in the "DATA" section.
-
CDC/Genome nCov data via China:
- nCov Positive Data Source/GenBank → "GenBank/Nucleotide/nCov search → Genbank/Usa/Wuhan seafood nCov complete genome
- nCov Positive Data Source/GISAID
- nCov Negative/Healthy Genome Data Source/1000 Genomes Project
- Using the genomics data above, may be suitable for ai based genomic analysis, via genomic deep learning.
- Suggested genome deep learning library, which deals with data in Genbank/FASTA format: Janggu
By extension, the tool by researchers at John Hopkins University below, is useful for real time tracking of nCov:
https://gisanddata.maps.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6
Note that despite the ~900+ infection-case number reported via China on January 24, by stark contrast, a medical scientific paper estimated that ~105,000+ infections actually occured at that time. All sources thus far ought to be scrutinized, as is typical in scientfic endeavour.