(1) The problem Covid-19 spreads before current tests diagnose it. Lockdown of hundreds of millions of people in the EU is therefore required to prevent its spread and resulting overwhelming health crises. But such lockdowns cause unsustainable social and economic impacts. There is a huge amount of data collected by doctors and hospitals worldwide which, taken together, may contain unexpected indicators of disease and treatment. This applies to Covid-19 (and other pandemics) on an individual, local and group basis. For example, do standard blood tests contain information that would allow us to diagnose Covid-19 prior to hospital admission, and improve treatment? Could we identify emerging clusters to catch new waves before they spread, to target social isolation and lockdown to specific areas, to prevent the necessity of doing these measures nationally? Could we do this for future Covid mutants, and other viruses and strains that tests do not currently exist for?

(2) Our Solution Our solution is an AI based on a special convolutional neural network which inputs available data about patients, their test results and, if required, their health records, geolocation etc, and outputs probabilities of specific outcomes, progression of disease, success of treatment etc. This week, in its first access to Covid-19 data, it found that in a decreased population of leukocytes, monocytes are predominant in Covid-19 patients. This is a novel finding for Sars-Cov-2 and significant for Covid-19 detection, and one we are submitting for peer reviewed publication right now.

(3) What you have done during the weekend We have worked on the core technology, making this the first time in the world that the power of Quantum computing and AI have been brought to bear on COVID-19, with the ability to handle data for hundreds of millions of people. We have developed the ability to distinguish Covid-19 from 12 similar virus infections. We have developed the ability to diagnose Covid-19 much earlier. We have progressed turning the system to studying the role of adjuvant immunity in the possible prevention of pandemics. We have worked with mentors Marco Venturelli (CFO Novartis Italy) and Mariano Guzzetta (Senior consultant EY) on a business plan to roll our solutions out over the EU, to allow member states to spot new waves and outbreaks before they develop, and allow localized lockdown and social isolation, rather than waiting for the recurrence to become national.

(4) The solution’s impact to the crisis (a) Separation of Covid-19 from non-Covid-19 patients, to reduce spread of infection, (b) Early treatment intervention (c) Clearer direction on appropriate pharmacological interventions. (d) Ability to lift lockdowns of hundreds of millions of people because of much earlier diagnosis, allowing patients to isolate before developing symptoms, and so bring the reproduction number to less than 1. (e) Impact is increased by our ability developed this weekend to distinguish Covid-19 from 12 similar viruses. (f) Ability to work over the whole of Europe: our move this weekend to Quantum AI means we can handle data on a massive scale, e.g. for everyone in Europe, and with a much larger number of data inputs per person, adding data such as genetic information, medical history, ICD-10 codes and additional data, allowing the AI to spot additional correlations that lead to health, social and economic solutions that impact hundreds of millions of people.

(5) The necessities in order to continue the project The initial findings from our winning hackathon last week are already being prepared for peer reviewed publication and fast track Covid-19 project grants in the UK, to allow the solution to be applied there. Suitable collaborators and funding are needed to roll out our predictive solutions across the EU, to allow us to prevent future waves of Covid-19 spreading. We will continue working with collaborators with whom we have advanced the project during this weekend’s hackathon, including Head of Laboratories at Centro Diagnostico Italiano (a large diagnostics group), a Swiss EHR company covering 35% of national records, Marco Venturelli, CFO of Novartis Italy, Mariano Gazetta, Senior Consultant EY Italy, consulting directly for Italian cabinet ministers.

(6) The value of your solution(s) after the crisis After the crisis our solution will prevent new waves of COVID-19, including mutants, which may not be picked up by current antibody tests, by bringing the reproduction number to less than one. It will also be able to detect and help isolate centres of future pandemics, as well as finding other information that is hidden in medical big data. In addition, by enabling lockdowns to be lifted our solution will reduce the negative economic impacts of the current crisis and so allow faster recovery after the crisis.

URL to the pitchvideo https://youtu.be/tw4xNBj3ntc

Built With

+ 5 more
Share this project:

Updates

posted an update

Discussions with team member Dr Louise MacKenzie has raised the exciting possibility of investigating adjuvant immunity (which existingr vaccines protect against COVID-19 to a certain extend), using the Athena AI

Amazing work by the whole team, with a special call out for Dr Abhirup Banerjee, who fuelled on Domino's pizzas and Diet Coke worked literally around the clock to move the codebase to make this the first Quantum AI to be addressing the Covid-19 pandemic!

Thanks also to our science team Dr Johanna, Dr Louise, Dr Bart, Dr Mike and Prof Surajit for a huge range of findings over the weekend including ensuring that the system distinguished COVID 19 from over 12 other symptomatically similar viral infections.

Many many thanks to mentors Marco Venturelli (CFO Novartis) and Mariano Guzzetta (Senior Consultant EY) and Oksana Bober for invaluable help and support developing and validating the business plan and introducing us and our project to a wonderful variety of contacts and connections from major test labs to putting it before several cabinet ministers.

And of course Shelley Taylor who PMed for us all and kept everything together.

Log in or sign up for Devpost to join the conversation.