The main objective of this project is to design and evaluate deep learning models for the automatic detection and classification of blood cancer from microscopic images. The specific objectives of the study are as follows: 1. Data Collection: Gather a diverse set of 652 labeled blood smear images, categorizing them into Benign, Pre, Pro, and Early stages of blood cancer. 2. Data Preprocessing: Apply necessary preprocessing steps, including resizing, normalization, and augmentation, to prepare the data for model training. 3. Model Development: Build four distinct deep learning models—Custom CNN, VGG16, MobileNet, and SqueezeNet—using appropriate architectures and techniques suitable for image classification tasks. 4. Model Training: Train the models on the preprocessed images and fine-tune them for optimal performance. 5. Model Evaluation: Evaluate the models using performance metrics such as accuracy, precision, recall, and F1-score on a test dataset to assess their classification capability. 6. Performance Comparison: Compare the performance of the four models and determine which model provides the most accurate and efficient results for blood cancer detection. 7. Best Model Selection: Select the best-performing model for possible integration into real-world diagnostic systems. Through these objectives, the project aims to explore the potential of deep learning in automating blood cancer detection and providing efficient solutions for medical diagnosis.