Creating any kind of robotic automation in surgery is risky. All sensors used have to be highly adaptable, ensuring that even the smallest amount of light variance doesn’t create some artefact for the image being processed. As a member of the robotics and autonomous systems research team, Dr Ajay Pandey leads the implementation of the novel image sensors and camera technology to solve this issue. Unlike current medical imaging tools which can only track the position of bone and medical tools, the system designed by the interdisciplinary robotics team combines advanced functionalities in pixel design to create state-of-the-art miniaturised cameras for robotic vision.
Project Background and Objectives
A theoretical approach to circumvent the vision impairment of current image sensors for machine vision was presented by Finlayson in which the use of a combination of narrow spectral responses and a logarithmic pixel was proposed to create an image that is invariant to the change in lighting conditions [G. Finlayson and S. Hordley, J. Opt. Soc. Am. A, 18, 253-264 (2001)]. Along these lines, in collaboration with researchers at Oxford University, we have shown that a combination of four spectrum selective sensors is capable of producing high purity of colour information in an image. Four narrow absorbers separately absorbing blue, green, yellow-orange, and red can produce images of natural scenes with high colour purity. Above all, the use of organic semiconductor as the sensing material allows to design intelligent cameras that can take any shape and form.
We aim to develop colour and depth sensing cameras for next generation of smart and flexible endoscopes for a range of minimally invasive surgical procedures, we’re going to do this by finishing the following objectives:
- Create voltage controlled Multispectral pixel operation
- Reversible pixel design for light emission and sensing
- Pixels as optical force sensors
- High sensitivity pixels for low level light detection and night vision