Cancer detection using machine learning has emerged as a promising approach for improving early diagnosis and patient outcomes. This research focuses on applying advanced algorithms such as Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and ensemble models to analyze medical imaging and histopathological data. The system automates feature extraction and classification, enhancing diagnostic accuracy and reducing human error. Data from breast, lung, and oral cancer datasets were used for model training and validation. Preprocessing techniques were applied to ensure image clarity and consistency. The proposed model achieved high precision and recall in identifying cancerous patterns. Limitations include data imbalance and interpretability challenges. Future work aims to integrate real-time diagnostics and multi-modal data for broader clinical use.