Leukemia is a life-threatening hematological malignancy that requires early and accurate diagnosis to improve patient outcomes. Manual examination of microscopic blood smear images is time-consuming, subjective, and highly dependent on expert pathologists. With recent advances in artificial intelligence, deep learning has emerged as a powerful tool for automated medical image analysis. The goal of this research paper is to develop a deep learning-based model that can accurately detect leukaemia from medical images, with a focus on optimizing the model's performance using advanced techniques such as transfer learning, hyper parameter tuning, and regularization methods. Evaluation metrics such as accuracy, precision, recall, F1 score, and the ROC-AUC curve will be used to assess the model's diagnostic ability. By building a robust and scalable deep learning model for leukaemia detection, this study aims to contribute to the growing body of research on AI-driven medical diagnostics and provide a practical tool to assist healthcare professionals in early leukaemia diagnosis.