Skip to main content

AI models for skin conditions management and diagnosis

Primary supervisor

Yasmeen George

Problem:

  • Almost 1 million people in Australia suffer from a long-term skin condition. 
  • Without early intervention, skin conditions become chronic conditions with significant health, psychosocial and economic impacts, including anxiety, depression and social isolation.
  • Access to safe, timely, high-quality specialist care leads to better outcomes for individuals.
  • With roughly 2 dermatologists per 100,000 Australians, it’s not surprising how hard it is to have access to dermatologist expertise.

Solution:

We aim to develop AI model that is equipped with specialist expertise to help improve diagnosis accuracy at primary care and offer optimal use of scarce health system resources. The model will be trained using skin images (clinical and/or dermoscopic) to identify disease relevant features and accurately diagnosis various skin condition/s.

Relevant resources: 

DOI: https://doi.org/10.1007/978-3-031-43987-2_20

DOI: https://doi.org/10.1007/978-3-031-43907-0_54

Required knowledge

Python, PyTorch, OpenCV


Learn more about minimum entry requirements.