Welcome to this specialization on AI for medicine. If you've completed the deep learning specialization or the machine learning course and you're looking for application areas to deepen your mastery of AI, this is a good specialization to take. One of the most important things for becoming really good at machine learning is to gain practice applying machine learning to multiple use cases. This specialization will step you through multiple use cases spanning some of the most important applications of AI to medicine. For example, given an image of a chest x-ray, so unstructured image data, can you train a neural network to diagnose whether or not a patient has pneumonia? You learn to do that in this specialization. Or, given structured data, such as a patient's lab results, can you train a decision tree to estimate the risk of heart attack? You learn to do that too. By working on these concrete problems, you also see a lot of the practical aspects of machine learning, from how to deal with imbalanced data sets, to how to work with missing data, to picking the right evaluation metric. In machine learning, we often default to classification accuracy as the metric, but for many applications, that's not the right metric. So how do you choose a more appropriate one? Even if your current work is not in medicine, I think you'll find the application scenarios and the practice of these application scenarios really useful, and maybe this specialization will convince you to get more interested in medicine. If you are interested in medicine, then this is a great specialization to take. AI for medicine is taking off all around the world right now, so this is actually a great time for you to jump in and try to have a huge impact. Maybe you can be the one to invent something that saves a lot of patients' lives. Let's get started. I want to introduce you to Pranav in the next video. Welcome back. I'm excited that this specialization will be taught by Pranav Rajpurkar, with whom I've had the good fortune of collaborating for several years now on AI for medicine research. Thank you for the intro. It's been wonderful working with you. This is a three-course specialization, and in the first course, you learn about building machine learning models for diagnosis. Diagnosis is about identifying disease. In the first course, you will build an algorithm that will look at a chest x-ray and determine whether it contains disease. You'll also build another algorithm that will look at brain MRIs and identify the location of tumors in those brain MRIs. So whereas the first course is on diagnosis or identifying disease, the second course will be on predicting the future health of a patient, which is called prognosis. In the second course, you will learn how to work with structured data. So let's say you have a patient's lab values and their demographics, and use those to predict the risk of an event, such as their risk of death or their risk of a heart attack. Finally, in the third course, you learn about AI for treatment, that is for the process of medical care, and also for information extraction, getting information out of medical text. In course three, you will learn how to use machine learning models to be able to estimate what the effect of a particular treatment would be on a patient. You'll also learn about the application of AI to text for particular tasks like question answering and for extracting labels from radiology reports. In this first course of the AI for medicine specialization, you learn about the applications of AI for medical diagnosis. Diagnosis means the process of determining which disease or condition explains a person's symptoms, signs, and medical results. In particular, you'll be learning how to build and evaluate deep learning models for the detection of disease from medical images. Just in the first week, you build a deep learning model that can interpret chest x-rays to classify different disease causes. In the second week, you'll implement evaluation methodologies to assess the quality of your model. And in the third week, you use image segmentation to identify the location and boundaries of brain tumors in MRI scans. Let's get started!