Chapter 8 Discussion With this explorative study we revealed in vivo wide‐band diffuse reflectance spectra (350 – 1830 nm, with 1 nm spectral resolution) of six tissue types within the human abdomen: colon, muscle, artery, vein, ureter and mesenteric adipose tissue. These spectra covered silicon (Si) and indium gallium arsenide (InGaAs) detector ranges, thereby exceeding wavelength boundaries (1600 nm respectively 1700 nm) reported by preceding work22,23,37,38. Spectroscopic measurements on human skin samples in the wavelength range of 1000 – 2200 nm have been reported previously39, but not regarding any of the tissues included by our study. Previously we reported DRS on freshly extirpated human colonic specimens, where vascular filling and oxygenation were disrupted26. The present in vivo study eliminates this shortcoming. Although the acquired reflectance spectra (Figure 8.3) show some similarities between the different tissue types, we reached promising results for automated tissue‐specific classification (ureter and artery from surrounding adipose tissue) in both sensor ranges (i.e., Si and InGaAs). Based on the classification accuracies (Tables 8.2 and Table 8.3) we can conclude that Si and InGaAs sensors are equally suited for automated discrimination between ureters and surrounding adipose tissue. Si sensors seem better suited for differentiation of arteries from mesenteric adipose tissue than InGaAs sensors. For both classifications (i.e. ureter‐adipose and artery‐adipose) in the Si sensor range, the blood‐related features (Figure 8.5, respectively Ft6 and Ft4) demonstrate the best results. Within the InGaAs sensor range, water‐related and fat‐related features (Figure 8.6, respectively Ft35 and Ft13) classified ureter and artery most accurately. The features used for classification are potentially significant due to differences in chemical composition (e.g., water/lipid content) of the investigated tissues and structures. Near‐infrared fluorescence imaging using exogenous contrast agents has been demonstrated as useful for real‐time, intraoperative visualization of ureters7‐9 and arteries10‐12. In our study the diffuse reflectance spectra were the basis to extract gradient and amplitude difference features. These spectra originate from intrinsic tissue properties (i.e., endogenous chromophores) that do not require contrast administration. Consequently, there are no problems with potential toxicity or allergy to a contrast agent. Future studies, comparing optical techniques based on either exogenous contrasts (e.g. near‐infrared fluorescence imaging) or endogenous contrasts (e.g. diffuse reflectance spectroscopy) contrasts, are necessary to disclose the clinical applications for these techniques. 120
proefschrift_Schols_SLV
To see the actual publication please follow the link above