ROBOPLAN/FANBOT: Federated AI-guided in-situ fast-track histology
Research type
Research Study
Full title
ROBOPLAN: Robotic and AI-assisted label-free multi-modality intraoperative tumour management with plane polarised light and micro-spectro-endoscopy. FINEBOT: Federated AI-guided in-situ fast-track histology using nano-precision robotic assisted triple modality planning during brain surgery.
IRAS ID
326961
Contact name
Joseph Davids
Contact email
Sponsor organisation
Imperial College RGIT
Duration of Study in the UK
3 years, 2 months, 1 days
Research summary
The present study aims to measure the interactions between different modalities used to record activity, morphology and behaviour of the cells within the brain. Plane polarised light microscopy, Raman spectroscopy analysis, probe-based confocal laser endomicroscopy (p-CLE), and probe-based continuous electrophysiology recordings will be used to measure brain tumour margins during their removal in theatre or from a tissue bank. These optical, spectroscopic and electrical measurements are routinely used clinically.
The mueller polarimeter is a method that looks at how light is oriented, whereas p-CLE device uses normal visible light to measure the way cells behave and displays them as an image that a histopathologist can read which is similar to what is seen when tissue is viewed under the microscope. Raman generates a signal as a graph that helps us identify constituent chemical bonds found in the tumour.
Electrophysiology is also routinely performed in theatre, because of the need to identify functioning tissue by measuring electrical activity of firing neurons, but does not penetrate the brain while the surgeon takes out the tumour.
The electrophysiology probe size is less than a millimeter, the p-CLE probe and the Raman probes are about 1.5mm in diameter. Because their size makes them challenging to manoeuvre, the surgeon can hold the tools using a probe-holder. In-vivo: once each of these probes are in-contact with the brain surface they will measure information over a period of 60-seconds, which will be sufficient for our analysis. Continuous recordings will be performed in accordance with the intra-operative brain tumour removal approach determined by the operating surgeon. It will not disrupt the surgeons workflow. Ex-vivo: once the surgeon has removed the tumour a robot will be used to perform a detailed scan of the tumour margins. The combined information the modalities provide will then be analysed using machine learning.
REC name
East Midlands - Leicester South Research Ethics Committee
REC reference
23/EM/0225
Date of REC Opinion
8 Nov 2023
REC opinion
Further Information Favourable Opinion