Deep Learning for Clinical Image Denoising

Primary supervisor

Jackie Rong

Co-supervisors

Convolutional neural network (CNN) has exhibited its significance in addressing large-scale vision tasks such as action recognition, image classification, super-resolution and denoising. Recently, many researchers have reported the potential of employing deep learning techniques to refine high-quality clinical images for diagnosis and treatment. Candidates in this research project will have the opportunity to work with real-world clinical data sets to develop novel deep learning methods to solve the problem by learning the relationship between clean and noisy input images. The resulting high-quality images will be further used for specific disease diagnosis.

Student cohort

Double Semester

Aim/outline

The aim of this research project is to design and develop deep learning algorithms to reduce the noise in clinical images for better diagnosis decision-making.

  • Exploring real-world clinical data and extract appropriate features.
  • Develop novel deep learning models to address the image denoising problem.
  • Use the output image to perform clinical diagnosis.

Required knowledge

  • Students have knowledge about machine learning and deep learning.
  • Student familiar with Python, R, or MATLAB.
  • Experience with data processing and image processing is preferred.