Young Jamaican develops world’s 1st known Artificial Intelligence Open source Covid19 Diagnosis tool -Call for Ministry of Health to deploy/utilize FDA-Approved Artificial Intelligence Diagnosis to detect Covid19

God Bennett, lecturer of “Universal Ai Diploma” with focus on (Neural networks), and creator of Robotize Jamaica in 2018, (non profit to improve Ja’s economy through Ai services) urges hospitals/medical institutions to start using CT scan based artificial intelligence algorithms, including recently fda-approved artificial intelligence based covid19 diagnosis solutions on May 11, 2020 (see Aidoc), also similar to the world’s 1st known open source covid19 diagnosis Ai tool, that Mr. Bennett had prepared back on February 9, 2020 as documented on Github. This tool of Bennett’s is now being incorporated into Adalabs Africa’s health platform.

There is a list of over 200 similar ai covid diagnosis tools seen in quick references at the article’s end, starting chronologically with Mr. Bennett’s ai tool, for interested parties.

Additionally, this could be used in concert with Jamaican doctors to detect Covid19, as a way to try to help prevent/control its spread, while providing adjunct diagnosis alongside the aforesaid overworked doctors and nurses.

80 Trillion dollar Ai market Cap predicted b Ark Invest by 2030

As Mr. Bennett lectures the Universal Ai Diploma, with focus on Applied Neural Networks, Neural networks are seen in applications from self-driving cars to disease diagnosis algorithms, and typically several aspects involving human thinking, which is many many aspects, especially where data is plentiful. These neural networks and other similar forms of Ai, is predicted to yield an 80 trillion dollar market cap by 2020. As Bennett underlined in 2018 local newspaper, Ai can aid economic stimulation especially when data is plentiful.

China & Use of Ai for detection, as early as February 20, 2020!

On February 20, 2020, news reports mentioned that China was using artificial intelligence to do Covid19 diagnosis, with about 98% accuracy, from CT scan images of lungs.

On Feb 9, 2020, intriguingly, Mr. Bennett had released an open-source artificial intelligence/machine learning based project for Covid19 detection, an alternative to the dna based polymerase method being employed by CDC. In the Ai field, it is typical to take a “pre-trained” Ai brain that accomplishes similar goals to the “new” desired task, similar to how an experienced driver is likely a better hire to do a new driving task over a human without driving experience. The Covid19 Ai diagnosis algorithm Mr. Bennett released, is trained or built on top of priorly existent pneumonia detection artificial intelligence software.

Pneumonia ai code was used as a basis, because as noted in my academic covid19 diagnosis paper, Mr. Bennett had discovered that for a large margin of cases, Covid19 shared many non-trivial similarities with pneumonia, as seen for eg in work by Zheng et al. For the task of non-Covid19 pneumonia detection, this code base had an overall Sensitivity/Specificity/Accuracy of ~89%, ~88%, and ~89% respectively.

On February 19 2020, a study showed that ct based radiology/diagnosis by human radiologists, of Covid19/”Sars Cov2" for detection, was roughly 98% accurate compared to the CDC proposed dna real time polymerase based method (roughly 71%). which further validates the use of Artificial Intelligence to aid in this Covid19 spread minimization task, with comparable performance to human radiologists or greater at 98%+ accuracies, enabling both medical and non-medical staff to do adjunct diagnosis, initially overseen by experience radiologists, therein freeing up medical staff for other core tasks.

Division of Artificial Intelligence Based Health Development is required!

In future scenarios, a “Division of Artificial Intelligence Based Health Development” or sector of artificial intelligence based research should reasonably exist in the Ministry of Health, that could enable Ai solutions to be rapidly researched/developed, to facilitate production of vaccines, and treatment, as seen in a recent example where MIT developed antibiotics based on Ai research/development, apart from the prior suggestion merely utilizing the aforesaid Ai health tools. This is also reasonably why Jamaica needs a “Minister of Artificial Intelligence”, as Mr. Bennett had pointed out in a 2019 Gleaner article.

Quick and easy references:

1. World’s first known Open-Source Machine Learning → Convolutional Artificial Neural Network in Covid19 XRay Diagnosis (February 9 2020), Submission for “65th Annual Health Research Conference, organized by the Government of Jamaica and Caribbean Public Health Agency” (Mr. Bennett)

2. Aidoc (Prof Paul Parizel et al, FDA Approved Covid19 Diagnosis Application)

3. Chinese Covid19 Diagnosis Tools and Academic papers: “Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography: a prospective study” (Jun Chen et al)

4. List of over 200 open source Covid19 Artificial intelligence tools, starting chronologically with Smart Covid19 Xray tool by Mr. Bennett:

5. Mr. Bennett’s covid tool in sample action:

Author: God Bennett, Artificial Intelligence Lecturer of “Universal Ai Diploma” at Advanced Solutions Technical Institute, Author of NVIDIA featured Ai pothole detector, creator of Cryptosynth Aeon conversational Blockchain Ai, creator of Robotize Jamaica, (non profit to improve Ja’s economy through Ai services) and Manager of Artificial Intelligence at Simble created by Erica Simmons Chairwoman Jamaica IEEE, and and Manager of Artificial Intelligence at Adalabs Africa



Lecturer of Artificial Intelligence, and inventor of “Supersymmetric Deep Learning” → Github/Supersymmetric-artificial-neural-network

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
God Bennett

Lecturer of Artificial Intelligence, and inventor of “Supersymmetric Deep Learning” → Github/Supersymmetric-artificial-neural-network