A technical presentation about processing medical images stored in DICOM format before passing the data in DL algorithms. DICOM files store image data along with lot of metadata. This presentation will also discuss the importance and usage of metadata.
Medical image processing is one of the most promising future of AI. For medical imaging scientists get data from CT Scans and MRI machines in DICOM format. These images are 3D images contains patient and image metadata along with image data. This metadata needs to be processed to get interested area from the image depending upon problem type. Processing includes:
1. Converting raw data into Hounsfield Unit by using slope and intercept values.
2. Understanding different DICOM metadata and how they are used during processing the data like window, level, slice thickness, slice increment etc.
2. Histogram Analysis.