Nicha Dvornek, part of Yale's Introduction to Medical Software course, walks through the basics of deep learning as a subfield of machine learning built on neural networks. She explains the computational model of a single neuron and how stacking layers creates a network, then moves through specific layer types: convolutional filters for image data, LSTM units for sequences, dropout for regularization, and pooling for downsampling. The lecture names concrete architectures, including the U-Net used in medical image segmentation, and surveys where these tools show up in healthcare, from radiology scans to genetic data analysis. It closes by naming popular deep learning toolkits students might use to implement these ideas. The pace is brisk and vocabulary-heavy, aimed at students who already know basic machine learning terms and want the deep learning layer explained in medical context rather than from first principles.