Image Analysis and Machine Learning for Medical Images
Medical images provide a glimpse of the internal state of the human body and relay information on the health and well-being of a subject. Radiology images, such as MRI, provide a gross description of disease, whereas digital pathology images describe disease on the cellular-level. Physicians must interpret these images, and render a diagnosis which is used for treatment planning. Unfortunately, human-based analysis is subjective, error-prone and inefficient, which ultimately reduces the quality of care for the patient. At the Image Analysis and Medicine Lab we are developing algorithms for neurological MRI and breast cancer digital pathology images to increase objectivity and efficiency of image interpretation to improve patient care. The student will be involved in software management, data management and algorithm development activities on these modalities.
The student will work on his/her own project, as well as with the other students to learn about image analysis and machine learning in medical imaging. The first part of the project will be around managing and organizing neuroimaging and pathology data. The intern will learn about the DICOM image standard, how to save data in a structure that allows processing en-mass, as well as how to load and process MRI/pathology data in batch processing mode using Matlab or Python. The student will also help to develop coding best practices and version controlling through github. They will get used to using some existing code, familiarizing them self with some algorithms, and working to make improvements to these algorithms.
Signals and Systems I and II
April Khademi : Image Analysis and Machine Learning for Medical Images | Friday March 22nd 2019 10:32 AM