Machine Learning Detection of Antimicrobial Resistance in Aquatic Bird Habitats

Abstract

In addition to their importance in maintaining ecological balance, wetland birds help to recycle nutrients as well as act as indicators of ecosystem health. Nevertheless, the presence of increased resistance of pathogens in wildlife has been identified as one of the significant issues facing the environment. This is caused by the possible movement of resistant microorganisms from the aquatic environment, through the animals, and finally to the nearby community members. Traditional microbial monitoring techniques are laboratory-based and hence time-consuming, expert-oriented, and do not offer a way for ecological monitoring. This study presents an alternative solution for detection of AMR risk in aquatic bird habitats through a machine learning algorithm that utilizes various attributes. Environmental factors like water quality, pollution, human impact, and wetland condition, in addition to biological factors such as bird population size, feeding habits, and migration pattern are included in the suggested model for risk prediction of AMR. The findings show that machine learning models are able to detect the complicated connection between the conditions of the habitats and AMR prevalence, with the proposed model having better performance than traditional methods. According to feature analysis, water pollution, human interference and microorganism exposure conditions were the main factors contributing to AMR risk. This study presents a useful framework for assessing wildlife health as it allows early identification of the most hazardous water habitats for microorganisms