AI uses for Medical Imagery
Before starting this course you must at least have an intermediate level of python, basic understanding of convolutional neural networks, and basic knowledge of Tensorflow. By the end of this course you will learn how to train very accurate convolutional neural networks to predict test images for binary class. You know enough to where if you want to go off on your own and use your own methods how to do that. Also appropriate parameters to use as well as data augmentation methods. It is explained in this course how to train multiclass as well. Not to mention you will learn how to use CV2 when predicting an image after training the convolutional neural network. You will also learn how to train a multi class Convolutional Neural Network and predict as well. Then learn to use a Keras Load Model Function for both binary and multi class predictions. Although the videos are short they are thoroughly and simply explained. You will also learn to deal with some of the challenges in deep learning as well when it comes to small dataset size. All the datasets featured in this video are found on Kaggle, except one that I provide to you directly. I will explain why in that video. Do not worry about the quizzes if you pay attention you will easily do great. But most importantly be ready to learn. This is not is challenging as it seems. I show you how to prevent overfitting and reduce bias severely with these methods in these videos.
What you’ll learn
- Convolutional Neural Networks.
- Data Augmentation.
- Tensorflow.
- Binary Class Predictions with percentages of each class.
- Keras.
- Maxpooling.
- CV2.
- AI uses for Medical Imagery.
- Creating a Train and Validation Set when the original dataset only has one folder.
- Dealing with Overfitting.
- Dealing with a very Small Dataset.
Course Content
- Introduction –> 2 lectures • 11min.
- Section 2 –> 3 lectures • 9min.
- Multi Class use cases in agriculture and Keras Load Model Function –> 3 lectures • 11min.
Requirements
- Intermediate Python Knowledge.
- Basic Tensorflow Knowledge.
- Either have an Google Colab. Or if using Jupyter Notebook Tensorflow and Keras already installed in the virtual environment.
- Have a basic idea of convolutional neural networks.
Before starting this course you must at least have an intermediate level of python, basic understanding of convolutional neural networks, and basic knowledge of Tensorflow. By the end of this course you will learn how to train very accurate convolutional neural networks to predict test images for binary class. You know enough to where if you want to go off on your own and use your own methods how to do that. Also appropriate parameters to use as well as data augmentation methods. It is explained in this course how to train multiclass as well. Not to mention you will learn how to use CV2 when predicting an image after training the convolutional neural network. You will also learn how to train a multi class Convolutional Neural Network and predict as well. Then learn to use a Keras Load Model Function for both binary and multi class predictions. Although the videos are short they are thoroughly and simply explained. You will also learn to deal with some of the challenges in deep learning as well when it comes to small dataset size. All the datasets featured in this video are found on Kaggle, except one that I provide to you directly. I will explain why in that video. Do not worry about the quizzes if you pay attention you will easily do great. But most importantly be ready to learn. This is not is challenging as it seems. I show you how to prevent overfitting and reduce bias severely with these methods in these videos.