About | Features | Technologies | Author
The computer vision possesses a broad range of applications in use for a great variety of domains, one of which is the healthcare-related medical industry. Artificial intelligence is now transforming healthcare. Besides computer vision, AI healthcare companies make good use of traditional machine learning algorithms and natural language processing as the cutting-edge tool-kits with greatest potential so as to explore everything ranging from drug chemistry to generic markers. For customer services, these companies are offering online consultants with predictive analytics, and meanwhile, they are incorporating test results and sensor data gathered from current patients in order to provide more and more real-time status updates to medical practitioners. As mentioned previously, the most exciting areas for AI in healthcare, are around computer vision and natural language processing.
This is a classifation based project build in the Fast.Ai Framework. The CNN (ResNet, DenseNet and Alexnet takes a x-ray image, and predict if the image belongs to a particular class.
✔️ Load images from a hierarchical file structure using an image datagenerator;
✔️ Apply data augmentation to image files before training a neural network;
✔️ Build a CNN using Fast.Ai;
✔️ Visualize and evaluate the performance of CNN models;
✔️ Take advantage of pretrained networks;
✔️ Understand what "freezing" and "unfreezing" a layer means in a neural network;
✔️ Implement feature engineering and fine tuning on a pre-trained model;
✔️ Compare various metrics like accuracy, precision, recall and AUC.;\
The following tool was used in this project:
Made with ❤️ by Muskaan Chopra