Participation at The Fourth Serbian International Conference on Applied Artificial Intelligence (AAI2025).

ESR3 Safi Ur Rehman Qamar participated in the Fourth Serbian International Conference on Applied Artificial Intelligence (AAI2025) which took place at Zlatibor, Serbia, on May 29th – 30th, 2025.

Safi presented his paper titled "Machine Learning Based Prediction of Cell Death After Silver Nanoparticle Treatment".

Abstract: Silver nanoparticles (AgNPs) are increasingly applied across biomedicine due to their antimicrobial and anticancer properties. However, concerns about their cytotoxicity to both cancerous and non-cancerous cells call for predictive tools that can accurately model their biological impact. In this study, we applied machine learning models to predict cell viability in response to AgNPs exposure using human umbilical vein endothelial cells (HUVEC) and breast adenocarcinoma epithelial cells (MDA-MB-231) as biological models. Experimental data were collected at 24 h and 72 h post‑treatment using a standard MTT assay. These data were used to train two machine learning models: Support Vector Regression (SVR) and Random Forest (RF). We then used each model to predict viability at 48 h and 96 h intervals not included in training and evaluated performance on unseen samples. SVR achieved RMSE of 1.61% at 48 h and 1.57% at 96 h outperforming RF (5.96% and 2.92%, respectively). Both models successfully generalized to unseen data and longer exposure durations. These results demonstrate the feasibility of using machine learning for nanotoxicology assessments enabling faster and more accurate predictions of cytotoxicity. These insights can significantly advance the safety evaluation of nanomaterials and support the development of safer nanoparticle based therapeutics.

Congratulations Safi!