AI-Based Predictive Modeling for Antibiotic Resistance in Bangladesh
Students & Supervisors
Student Authors
Supervisors
Abstract
Background: Antibiotic resistance has been worsening in Bangladesh for years. It wasn't tracked properly in the early 2000s. The available data was poor, and the methods were mostly manual. Now, the situation is much more serious. The resistance index was 0.20 in 2000, but it climbed to 0.85 by 2024. This study examines how AI tools began to predict this issue and whether they were effective. Methods: We used data from 2000 to 2024. Initially, the data was messy and not very useful. However, from around 2010, things improved. Some hospitals and health centers started testing AI models like logistic regression and random forest. We monitored the accuracy of these models based on the amount of data available. In cities, the data was solid, leading to better results. In rural areas, the data was lacking. Results: The accuracy of these models improved significantly over time. It was about 61% when AI tools were first used seriously, and now it exceeds 90% in some locations. Yet, rural areas still face challenges due to incomplete data. Additionally, the models performed best when there was abundant data and strong AI integration. This is not surprising, but it's still important. Conclusion: AI cannot magically solve antibiotic resistance, but it certainly helps. By collecting better data and training people in all areas, not just the major hospitals, we can make a real difference. This study shows that AI tools combined with proper data can help doctors make informed decisions before it's too late.
Keywords
Publication Details
- Type of Publication: Conference
- Conference Name: 1st National BioMed Health ResearchCon (NBHRC) 2025
- Date of Conference: 28/08/2025 - 28/08/2025
- Venue: Dr. Milon Auditorium, Dhaka Medical College
- Organizer: Dhaka Medical College Research and Academic Club (DMC-RAC)