Multi-modal optical imaging applied to robot-assisted endoscopy (MOIRE)

Lead Research Organisation: Imperial College London
Department Name: Institute of Biomedical Engineering

Abstract

Over the past fifteen years the strong and complex interaction of light and tissue has begun to be exploited in clinical applications for the diagnosis of disease. The absorption, scattering, reflection and fluorescence properties can provide a live signal to reveal metabolic and morphologic changes in the epithelium and stroma associated with the development of pre-cancer. The potential increase in diagnostic specificity through multi-modal and multi-spectral imaging may reduce the number of unnecessary biopsies and reduce the delay to intervention, improving the current slow and inefficient white light screening procedures. During the same time period, there has also been a revolution in surgical technology through the widespread use of minimally invasive surgical (MIS) techniques and the development of sophisticated human-controlled surgical robots, which allow the performance of procedures that are otherwise prohibited by the confines of the operating environment. Currently, the surgical navigation for these procedures is based on white light reflected video images acquired from conventional or stereoscopic endoscopes, giving a limited view of the tissue that only reports tissue colour, shape and texture. The purpose of this project is to develop a multimodal optical visualisation technique that permits in situ, in vivo imaging of complex anatomical and functional information of the operating field and integrate this into robot assisted surgery for safer and more effective surgical navigation and diagnosis.

Publications

10 25 50

publication icon
Wood TC (2010) Polarization response measurement and simulation of rigid endoscopes. in Biomedical optics express

publication icon
Wood T (2012) A tunable supercontinuum laser using a digital micromirror device in Measurement Science and Technology

publication icon
Wood TC (2008) Optimal feature selection applied to multispectral fluorescence imaging. in Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention

publication icon
Clancy NT (2012) Light sources for single-access surgery. in Surgical innovation

 
Description The absorption, scattering, reflection and fluorescence properties of biological tissue can provide a live signal to reveal metabolic and morphologic changes in the epithelium and stroma associated with the development of pre-cancer. The potential increase in diagnostic specificity through multi-modal and multi-spectral imaging may reduce the number of unnecessary biopsies and reduce the delay to intervention, improving the current white light screening. This project investigated the use of sensing and imaging modalities together with minimally invasive surgery (MIS) techniques and surgical robots. During this project, a multimodal optical visualisation technique has been developed that permits in situ, in vivo imaging of complex anatomical and functional information of the operating field and this was integrated with robot assisted surgery.



The results of this project are assessed against each of the original objectives below.



O1 To develop a computer controlled rapidly tunable excitation source that is able to deliver an arbitrary excitation spectrum to the sample.

This has been achieved using a Digital Multimirror Device and the results are were published in MST (T. Wood and D. S. Elson, Design, Construction, and Characterisation of a tunable supercontinuum laser using a DMD, Measurement Science and Technology 23 105204 2012).



O2 To adapt conventional laparoscopes for multimodal imaging with i) white-light, ii) excitation-resolved fluorescence and iii) excitation-resolved cross-polarized scattering.

This has been achieved and published (e.g. invited talk by T. Wood et al. Development and Analysis of a Polarised Endoscopic Hyperspectral Reflection and Fluorescence Imaging System, ECBO, SPIE/OSA, Munich, 2009 and journal article Polarization response measurement and simulation of rigid endoscopes, Biomedical Optics Express 1 463 2010). Beyond the scope of the project we have also adapted a da Vinci triple endoscope and used a computer vision method to allow multimodal image registration (e.g. D. S. Elson et al. Spectroscopy and spectroscopic imaging using surgical robots. OWLS. Quebec, 2010, and journal article Clancy et al. Multispectral image alignment using a three channel endoscope in vivo during minimally invasive surgery, Biomedical Optics Express 3 2567 2012). This work is being prepared for journal publication and has resulted in further industrial funding from Intuitive, who are also in discussion with Imperial about related IPR.



O3 To achieve meaningful contrast using parameters from the large multi-dimensional data sets acquired from in vitro samples. This will include testing and comparison with histopathology on superficial tumours of the pancreas, liver and colon, and on colonic specimens with ulcerative colitis and Crohn's disease.

Novel dimensionality reduction algorithms were developed based on IsoMap and BFFS and were published at MICCAI (T. Wood et al. Optimal Feature Selection Applied to Multispectral Fluorescence Imaging. MICCAI. New York: Springer 2008.) These have been tested on many tissues and results published.



O4 To incorporate the endoscope into the endoscope arm of the da Vinci robotic surgery system and interface the image data into the surgeon's console.

The endoscope was integrated under robot control although the image data was not routed to the surgical console since this would have required access to the da Vinci API software and hardware for video overlay. Instead the data was displayed using a laptop computer.



O5 To assess the potential of the technology in a clinical evaluation by engineering a prototype that can be evaluated by a surgical team practicing simulated tasks.

A prototype was constructed and evaluated both in a simulated environment but also in a large animal trial at Northwick Park Hospital. This was beyond the proposed aims (preliminary results reported e.g. D. James et al. 12th World Congress of Endoscopic Surgery. National Harbor, MD: SAGES, 2010, and V. Sauvage, et al. Advanced Biomedical and Clinical Diagnostic Systems, SPIE, SF, 2010. Paper published Sauvage et al. Multi-Excitation Fluorescence Spectroscopy for Analysis of Non-Alcoholic Fatty Liver Disease, Lasers in Surgery and Medicine 43 392 2011)
Exploitation Route Parts of the system have been adapted for in vivo data acquisition and may find applications in screening or image guided surgery. We are currently working with industrial and other academic partners towards the exploitation of the developed technology, including further funding from EPSRC and the EU.
Sectors Digital/Communication/Information Technologies (including Software)

 
Description European Research Council
Amount £1,376,510 (GBP)
Funding ID 242991 
Organisation European Research Council (ERC) 
Sector Public
Country Belgium
Start 12/2009 
End 09/2015
 
Description Intuitive Surgical Inc
Amount £35,633 (GBP)
Organisation Intuitive Surgical Inc 
Sector Private
Country United States
Start 08/2011 
End 05/2012
 
Description Intuitive Surgical Inc
Amount £35,633 (GBP)
Organisation Intuitive Surgical Inc 
Sector Private
Country United States
Start  
 
Description National Institute for Health Research
Amount £65,576 (GBP)
Funding ID II-3A-1109-10038 
Organisation National Institute for Health Research 
Sector Public
Country United Kingdom
Start 06/2010 
End 05/2011