Tissue Classification using CNN

• Built a CNN model to classify 100,000 tissue images from the PathMNIST dataset into 9 categories, achieving 87% accuracy on test data.

• Designed a deep learning architecture with 4 convolutional and 2 max pooling layers, utilizing the Adam optimizer and categorical cross-entropy for optimized training.

• Implemented detailed visualizations including prediction confidence scores and accuracy metrics, enhancing model interpretability and evaluation.

Tech Stack: Python, TensorFlow, NumPy, Matplotlib, Image Classification, Convolutional Neural Networks (CNN).