Interpretable Lung Cancer Detection using Explainable AI Methods
Md. Sabbir Ahmed, Khondoker Nazia Iqbal, Md. Golam Rabiul Alam · 2023
Lung cancer is the second most frequently diagnosed cancer worldwide. It is one of the major causes of death among both men and women. In order to reduce the death rate, early detection of lung cancer is the only approach. In this paper, we have proposed an interpretable lung cancer diagnostic system using different ML models like Decision Tree, Logistic Regression, Random Forest, and Naive Bayes classifier. A Kaggle dataset named ‘Lung Cancer Detection’ is used to perform our research work. Among different ML models, Logistic regression and Random forest classifiers scored the highest accuracy of 97% in predicting lung cancer. We have also used the two most popular XAI models, SHAP and LIME to show the interpretability of our used models.