Speaker: Daniele Ravi and Neil Clancy
UCL Contact: Dominique Drai (Visitors from outside UCL please email in advance).
Date/Time: 18 Oct 17, 13:00 - 14:00
Venue: Roberts 421
Title: Intra-operative hyperspectral imaging for brain tumour detection and delineation
Recent advances in hyperspectral imaging have made it a promising solution for intra-operative tissue characterization, with the advantages of being non-contact, non-ionizing, and non-invasive. Working with hyperspectral images in vivo, however, is not straightforward as the high dimensionality of the data makes real-time processing challenging. In this seminar, a novel dimensionality reduction scheme and a new processing pipeline are introduced to obtain a detailed tumor classification map for intra-operative margin definition during brain surgery. Existing approaches to dimensionality reduction based on manifold embedding can be time consuming and may not guarantee a consistent result, thus hindering final tissue classification. The proposed framework aims to overcome these problems through a process divided into two steps: dimensionality reduction based on an extension of the T-distributed stochastic neighbor approach is first performed and then a semantic segmentation technique is applied to the embedded results by using a Semantic Texton Forest for tissue classification. Detailed in vivo validation of the proposed method has been performed to demonstrate the potential clinical value of the system.
Title: Endoscopic Spectral Imaging
Optical absorption spectra of biological tissue contain clinically useful information related to its health. Chromophores such as haemoglobin, in particular, may be used to assess blood volume and tissue oxygen saturation within an organ, which gives an indication as to the health of its blood supply. Clinically, this can be used to guide surgical procedures, indicating areas of ischaemia, evaluating the success of organ reperfusion, and indicating areas of disease. We have developed systems for minimally invasive imaging in the clinic compatible with both rigid (laparoscopic) and flexible endoscopic platforms. The rigid system uses xenon illumination and a liquid crystal tuneable filter to switch between the wavebands that reach the camera. The measured absorption spectra at each pixel location are then decomposed into contributions from oxy and deoxyhaemoglobin, following an image registration step. The flexible system is a hyperspectral imager coupled to a fibre optic imaging bundle, which can then be used inside standard colonoscopes. Both systems have been trialled in animal procedures and used to measure oxygen saturation dynamics within the bowel and the uterus intraoperatively. A clinical version of the flexible imager has also been used to image human colonic polyps in a recent pilot study. Spectral imaging is proving to be a versatile tool at the clinician’s disposal, unveiling optical contrast and quantitative results that are otherwise invisible under standard imaging. Our studies have shown that it is possible to measure spatial changes in oxygenation longitudinally to follow the response of biological tissue to surgical provocation. Initial results have also demonstrated that this technique holds promise as a diagnostic instrument in the identification of abnormal tissue in minimally invasive procedures.