The integration of multi-modal Lung Cancer data and dimensionality reduction techniques has become a focal point in medical data analysis due to its potential to enhance diagnostic accuracy and predictive modelling. This research introduces a novel hybrid Feature High Dimensionality Reduction approach combining Principal Component Analysis (PCA), t-Distributed Stochastic Neighbour Embedding (t-SNE), and Linear Discriminant Analysis (LDA) for dimensionality reduction, followed by multi-modal feature integration using optimized fusion techniques. The study utilizes diverse medical datasets, including imaging and genomic data, to create comprehensive patient profiles. The integrated model employs standard deviation normalization to ensure equal contribution of all features, addressing issues of feature imbalance. The Hybrid Feature High Dimensionality Reduction (HFHDR) model integrates PCA, t-SNE, and LDA for dimensionality reduction, followed by multi-modal feature integration and standard deviation normalization. Initial experiments using PCA, t-SNE, and LDA yielded accuracies up to 0.88. Multi-modal feature integration strategies further enhanced accuracy, with Hybrid Fusion achieving 0.92. Standard deviation normalization in the final phase resulted in the highest performance, with an accuracy of 0.94 and an AUC of 0.96. These findings demonstrate that the HFHDR model effectively reduces dimensionality and integrates diverse data sources, significantly improving the analysis and prediction of lung cancer outcomes.