AIbolit is a project that aims to bring together machine learning methods for disease prediction.
Most medical datasets are unbalanced, for this reason AIbolit uses SMOTE in combination with RandomForestClassifier. In the ipynb file you will find the learning outcomes, correlation matrix, confusion matrix and other work steps.
Used datasets: https://www.kaggle.com/datasets/einsteindata4u/covid19 https://www.kaggle.com/datasets/fedesoriano/stroke-prediction-dataset https://www.kaggle.com/datasets/mysarahmadbhat/lung-cancer
There is an idea to create a single service for all available predictions based on clinical tests, so patients can independently determine whether they are at risk and consult a doctor.
