Antimicrobial Resistance Tracker

Antimicrobial Resistance is a leading cause of death for all ages. The AI model  will help to identify geographical patterns of AMR spread, predict future trends and track stewardship progress.

     As a result of antimicrobial resistance (AMR), microorganisms can persist or grow even when they are treated with drugs intended to inhibit or kill them. Microorganisms that become resistant to antibiotics are often untreatable, and in some cases, no drugs can provide effective therapy. As a result, treatments do not work. Consequently, humans, animals, and plants experience an increase in disease and mortality. A common cause of antimicrobial resistance is misuse or overuse of antibiotics.  

     Antimicrobial resistance has been identified as a silent pandemic because it spreads without us knowing its extent or severity. “AMR is sadly silent because it has not been discussed like other pandemics – and of course has been in the shadow of the COVID-19 pandemic in recent years”-Melissa Gong. Already, the drug-resistant diseases cause at least 700,000 deaths worldwide each year, and “if no action is taken,” that figure could increase to 10 million globally per year by 2050, overtaking diabetes, heart disease and cancer as the leading cause of death in humans, (8 Jul 2019). As a leading cause of death for all ages, It kills more people than HIV or malaria and has been identified by WHO as 'one of the top 10 global public health threats facing humanity.' Without action, the rise of AMR cumulatively may result in over 3.4 trillion USD loss in the world’s annual gross domestic product (GDP) in ten short years. 

I am proposing an Artificial Intelligence model that can estimate geographical trends of Antimicrobial Resistance (AMR) spread and predict future trends of AMR in specific locations, based on a slow or rapid rise in resistance. Changing treatment policies or limiting transmission of resistant bacteria may mitigate increasing antibiotic resistance in a location. Consequently, it would be advantageous to be able to predict accurately the distribution of antibiotic resistance.

To ensure a holistic view of this problem, data across multiple human health, agricultural and environmental sectors would be utilized. The dataset will be constructed from online databases showing various statistics on sales, usage and administration of antibiotics and deaths caused by AMR. Labels will include countries, cities and their potential risk levels. Information such as demographics and clinical profiles of patients would also be taken into account to measure effectively the burden of aMR. Adding antibiotic prescription or consumption data to the model would improve its predictive accuracy. Since AMR is communicable, addition of data that includes the rate of mobility to certain countries would also increase the accuracy of prediction. Mobility and antimicrobial resistance are complexly correlated. Recent findings suggest that various modes of mobility can contribute to transmission of these diseases.

The model would be able to:

 -Identify potential risk levels of each location (cities or countries)

-Point out places that need immediate or rapid action

-in turn, show the progress of AMR Stewardship, and

-Recommend AMR Stewardship practices that could be taken, based on data.

     The importance of this AI model is to serve as a resource system for agencies that formulate guidelines for treatment and prevention of AMR transmission to carry out antimicrobial resistance stewardship- antimicrobial stewardship programmes.

     The model would also come in handy as a learning tool also to students just coming into the field of AMR Research, because the best way to research VMR is to learn from data or trends. AMR Advocates and the government would also benefit from the predictive analysis. It would also contribute immensely to the fight against this potential pandemic. It is important to note that the application would not be used to make decisions on behalf of a human. Instead, it would act as an alert system. So the final decision is still made by humans.

My team and I are a set of young and motivated college students with enthusiastic spirits for research and delivery. I have sufficient knowledge to understand the concepts of artificial intelligence in antimicrobial resistance. Apart from studying biology, I have attended and gained knowledge in the aspects of artificial intelligence for human good. My team also consists of computer science and engineering majors with adequate knowledge in the aspects of data science and machine learning. 

Beyond the professional or educational aspect, we are passionate about putting our skills and knowledge to good use for the benefit of humanity. The implementation of a project like this could save many lives and be of benefit to humanity. 

We're constantly researching new methods on how to best implement our project. Although this concept is still in the idea phase, we've met with various professionals in different fields to talk about the plausibility and feasibility of our concept.