Deep learning for classifying cancer types using multi-omics data (Preprint)

BACKGROUND Cancer is one of the leading causes of disease burden globally, and early and accurate diagnosis is crucial for effective treatment. This study presents a deep learning-based model designed to classify five common types of cancer in Saudi Arabia: Breast, Colorectal, Thyroid, Non-Hodgkin Lymphoma (NHL), and Corpus Uteri. OBJECTIVE To determine whether incorporating multi-omics data, including RNAseq, mutation, and methylation data, could enhance classification accuracy. METHODS Utilizing a stacking ensemble learning approach, our model integrates five well-established methods: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Artificial Neural Network (ANN), Convolutional Neural Network (CNN), and Random Forest (RF). The methodology involves two main stages: data pre-processing (including normalization and feature extraction) and ensemble stacking classification. We prepared the data before applying the stacking model. RESULTS The stacking ensemble model achieved 98% accuracy with multi-omics versus 96% using RNA-seq, suggesting that multi-omics data can be used for diagnosis in primary care settings. The models used in ensemble learning are among the most widely utilized in cancer classification research. Their prevalent use in prior studies underscores their effectiveness and flexibility, enhancing the performance of multi-omics data integration. CONCLUSIONS This study highlights the importance of advanced machine learning techniques in improving cancer detection and prognosis accuracy. It contributes valuable insights by applying ensemble learning to integrate multi-omics data for more effective cancer classification. CLINICALTRIAL Not Applicable