AMR-ML

Antimicrobial resistance is a looming public health threat that requires urgent attention. To curb the emergence and spread of AMR, it is imperative to analyze resistance patterns from surveillance data to observe trends and inform clinical practice. In our study, we aim to utilize supervised machine-learning techniques on existing real-world surveillance data to help predict AMR trends across the world. Given the bacteria isolated, and antibiotics tested, along with other demographic details, the models implemented will be able to predict the resistance phenotype as susceptible or intermediate. In addition to our ongoing study, the updated datasets will be of great importance and will make the models more robust and accurate. These models could aid in informing public health practices and strengthen health systems with the help of state-of-the-art technologies such as machine learning.