<p>Lung adenocarcinoma (LUAD), the most common subtype of non-small cell lung cancer, remains a leading cause of cancer-related mortality worldwide. Early detection is hindered by the high dimensionality and noise of gene expression data, which complicate the design of accurate and interpretable diagnostic models. Existing LUAD detection approaches often rely on deep learning or traditional tree-based classifiers, typically sacrificing interpretability and handling biological uncertainty poorly. To address this, a framework is proposed in this work that integrates Genetic Algorithm (GA) based feature selection with Fuzzy Decision Tree (FDT) modeling for early LUAD detection. The GA module identifies compact, discriminative gene subsets, while the FDT constructs transparent fuzzy rule-based classifiers. The framework is evaluated on four widely used LUAD datasets and consistently outperforms classical Decision Tree, Random Forest, Support Vector Machine (SVM), and Logistic Regression baselines across accuracy, precision, recall, F1-score, and ROC-AUC. The resulting fuzzy rules provide clear decision logic, enhancing transparency and supporting clinical use. The proposed GA-FDT framework achieved accuracies of 91.67% on GSE19804, 99.15% on the TCGA-LUAD cohort, 98.42% on GSE32863, and 93.66% on GSE10072, with F1-scores up to 0.9929 and ROC-AUC values up to 0.9976. Across these datasets, GA-FDT delivers accuracy improvements of up to 6.83% and F1-score gains of up to 8.14% over GA-optimized Decision Tree, Logistic Regression, Random Forest, and SVM baselines, demonstrating robust and generalizable performance for LUAD detection.</p>