ELOUTMADI/Dicom-Images-preprocessing

Ultrasound imaging has become one ol the most widely used medical imaging modalities in clinical practice, it is characterized by low image quality with great variability compared to other modalities. Researchers have thus applied computer vision and deep learning techniques to ultrasound DICOM images for classification, detection and segmentation. These techniques, despite offering Interesting performances are very hard to implement at a large scale because of the complexity of the data coming from different sources and with different distributions. Fortunately, the cloud is changing the way applications are designed, including how data is processed and stored before training and deploying deep learning models. In this project, we aim to design and build a DICOM data pipeline in Microsoft Azure that describes how data flows through a solution, wherê it is processed, where it is stored, and how il is consumed by the next component in the pipeline including an annotation platform, a training cluster or a deployment endpoint.