| dc.description.abstract | Antibiotic resistance is a growing global threat that could potentially cause a future pandemic.
Machine Learning (ML) is becoming more common in medical research and healthcare,
improving clinical practices. One of the most important uses of ML is to fight Antimicrobial
Resistance (AMR), as rising resistance to antibiotics and the difficulty of treating multidrug resistant infections present serious challenges globally, with potentially life-threatening
outcomes. In the era of increase in AMR, it is critical to identify antibiotic resistance early,
raise public awareness about the issue, and implement comprehensive antibiotic stewardship
programs. These steps will help limit the misuse of antibiotics and prevent the AMR problem
from worsening. To develop an ML model, data from 224 patients in three public hospitals in
Bangladesh were collected, resulting in a dataset of 1,119 instances. The Random Forest
algorithm was used to evaluate antibiotic predictions based on patient prescriptions and clinical
data, with training and testing. Additionally, a public dataset was used to assess people's
knowledge, attitudes, and practices regarding antibiotics, processed using linear regression.
The data included information such as gender, age, marital, education, occupation, monthly
income, residence/area, religion in socio-demography. By applying the Random Forest
algorithm (a supervised classifier), the model achieved an accuracy of 0.90, indicating strong
predictive ability. In terms of public knowledge, linear regression analysis showed that 20.3%
of participants had very weak knowledge, 22.9% weak, 18.1% moderate, 18.9% strong, and
19.8% very strong knowledge of antibiotics. Supervised ML tools have proven effective in
predicting antibiotic resistance. This paper reviews how ML and Artificial Intelligence (AI)
can be used to predict antibiotic resistance and discusses how these tools can help public
improve antibiotic use practices | en_US |
