Despite the successful treatment of some forms of the disease, lung cancer continues to be the leading cause of mortality worldwide and occurs largely at an advanced stage because the disease is asymptomatic in its early development. This paper addresses this challenge by proposing a robust lung cancer diagnostic framework based on decision trees for early detection of lung cancer. First, they are more interpretability friendly for their interpretability in terms of the traceability of diagnostic pathways and usable insights. Preprocessing methods including ADASYN are used to overcome the class imbalances, along with a pruning technique to prevent overfitting while using demographic, clinical and symptomatic features to form the model. The decision tree was evaluated using accuracy, precision, recall and F1 score, the test data accuracy was 94%. The presented methodology is scalable, transparent and suitable for real world applications; as a result we present a solution for evaluating early lung cancer detection that is accessible and interpretable. The framework is then compared with related works for its reliability and adaptability, and challenges faced including imbalanced datasets, feature selection, clinical workflows integration. In this study we demonstrate the capacity for decision trees to democratize lung cancer diagnostics and enhance outcomes, particularly in underserved areas.