Predicting Antibiotic Resistance in Gonorrhea: A Comparative Analysis of Machine Learning, Deep Learning, and Explainable AI

Antibiotic-resistant Neisseria gonorrhoeae, the bacterium responsible for gonococcal gonorrhea, is an increasing public health problem. If left untreated, gonorrhea can result in complications such as infertility and pelvic inflammatory disease; the rising resistance to antibiotics complicates infection control. We performed our experiments with different models such as Logistic Regression, Decision Trees, Random Forests, and Support Vector Machine (SVM) over more complex Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM) and Gated recurrent units (GRU). We used Explainable AI (XAI) to lighten the black box and infer what decisions these models make. In general, deep learning models were more successful in capturing the data patterns, with XAI being a key mechanism to provide knowledge on operability that can guide the treatment strategies.