Machine learning has emerged in recent years as a powerful tool for many tasks across a wide number or disciplines. This has held true in biomedical imaging, where machine learning-based technologies have the potential to improve the efficiency and accuracy of imaging specialists by automatically identifying and measuring key findings within image data. Unfortunately, those automatic tools do not exist yet, and manual annotation is the common, time-consuming, standard. The purpose of this project is to develop a medical image annotation tool that will allow researchers to label medical imaging data in a facile manner and predict annotation in an automated fashion.
Check out the project demo website here.