Automated spectroscopic tissue classification in colorectal surgery As this study describes an initial exploration towards future spectral imaging guided surgery, providing an indication of the procedural costs of optical guidance during surgery, based on diffuse reflectance spectroscopy or hyperspectral imaging technique, is rather difficult at this moment. However, it is reasonably to expect that costs will be in the same range as already commercially available imaging systems for image‐guided surgery, such as near‐infrared fluorescence imaging. An advantage of endogenous contrast based imaging is that there is no need for administration of a contrast agent, which already results in a cost‐reduction compared to e.g. near‐infrared fluorescence imaging technique. Spectral imaging using absolute signal intensity to classify tissues22 might experience problems with robustness regarding variations in conditions in the operation room (e.g. operating theatre light intensity). We explored the spectral tissue signatures by extraction of relative spectral features from DRS signal which are quite robust with regard to variations in signal intensity27. Although the study sample was reasonable and justifiable in research setting (from a statistical point of view), we only describe experience in ten study subjects. From a practical perspective a more extensive study is needed to judge the full clinical relevance in daily practice. As the LOO cross‐validation method inherently produces relatively optimistic classification results, external validation remains essential before classification models can be implemented in clinical practice40. Such validation would need to be performed on newly acquired data in a multi‐center study. In this study, we use physiological knowledge on composition of different tissue types, and thereby use the pre‐defined spectral regions of interest covering water, fat and hemoglobin. Tissue differentiation on the basis of completely automatically extracted features from a larger data set might achieve better results, and should be explored in future work. Additionally, as the acquired reflectance spectra are specific to the probe geometry, more research is needed on the relation between these reflectance spectra and in vivo intrinsic tissue biological properties which potentially could be captured by optical tissue properties. This is needed to obtain a better understanding of the nature of discrimination performance: i.e. whether ureters and arteries can be optically distinguished from surrounding adipose tissues based on differences in light scattering behaviour (related to structural differences) or due to differences in absorption behaviour (related to chromophore concentrations such as blood, water and fat). Such an approach would also be tissue‐specific and robust to inter‐patient and multi‐center variability. The gold standard used in this study was the surgeon’s visual judgment. This judgment is based on color (spectral) information, on the recognition of anatomical position and 121
proefschrift_Schols_SLV
To see the actual publication please follow the link above