Intelligent Algorithm Development
Intelligent Algorithm Development
At the Photomedicine Labs, we research and develop intelligent algorithms that use machine learning and image processing techniques to improve the quality of the PARS modality through a holistic problem solving method by considering the entire optical imaging chain from the sample we are imaging, to the optical components, to the method in which we are capturing our signals, and to the way in which we process our signals and present them. In addition, these algorithms can be parameterised to include system and component properties that can be optimised over to iterate on the actual design of the PARS hardware allowing there to be a simultaneous and iterative co-design of algorithm and optical hardware.
Current research projects focus on:
At the Photomedicine Labs, we research and develop intelligent algorithms that use machine learning and image processing techniques to improve the quality of the PARS modality through a holistic problem solving method by considering the entire optical imaging chain from the sample we are imaging, to the optical components, to the method in which we are capturing our signals, and to the way in which we process our signals and present them. In addition, these algorithms can be parameterised to include system and component properties that can be optimised over to iterate on the actual design of the PARS hardware allowing there to be a simultaneous and iterative co-design of algorithm and optical hardware.
Current research projects focus on:
- Chromophore understanding and identification through linear and non-linear inverse models.
- Enhancing image quality and contrast through unsupervised learning with novel loss functions.
- Improving image understanding through virtual modality translation utilising generative networks