Figure 1.
Mechanism of hemoglobin oxygen saturation imaging and schematic illustration of prototype endoscope system.
(A) Illustration of the mechanism (see text for details.) (B) The 445-nm laser excited a phosphor equipped at the tip of the endoscope and emitted white light. The 473-nm laser light was emitted without the phosphor excitation. These two lights alternately illuminated the mucosal surface and the reflected lights were sequentially detected with a colour CCD in synchronization with light switching. The obtained images were processed and transformed into a StO2 map.
Figure 2.
Verification of hemoglobin oxygen saturation imaging by observing a phantom.
(A) Blood vessel phantom consisted of a glass tube filled with diluted blood and aqueous solution of intralipid. The intralipid solution strongly scattered incident light to simulate the living tissue around blood vessel. (B) The observed optical densities of the blood vessel at the three bands were dependent on StO2(T) (left) and Hct (right). Here, StO2(T) denotes the supposedly correct value of StO2 derived by analyzing the transmittance spectrum of blood. (C) StO2(I) map (derived by image processing) of the vessel. (D) Comparison of StO2(I) with StO2(T) (derived by measurement of transmittance spectra).
Figure 3.
In vivo imaging of nude mouse implanted with cancer cells.
(A) White light image of the mouse. The solid line corresponds to the cross-section of pathological assessment. (B) StO2 map of mouse before transplantation (left). Hypoxia developed at the tumor at 7 days after transplantation (right). (C) Histological picture (hematoxylin-eosin stained) of skin resected from the mouse at 14 days after transplantation (upper right, lower). Arrows indicate corresponding vessels.
Figure 4.
(A) X-ray fluoroscopic images during application of transcatheter arterial embolization in pig stomach. (i) Image of the target vessel. (ii) Image of the endoscope inserted into the stomach. (iii) Red arrow indicates marking clip to identify the target area. (iv) Image of injected hystoacryl medium into the artery from the catheter. (B) White light images (upper) and StO2 maps (lower) of the gastric mucosal surface visualized by laser endoscope system before embolisation (left) and five minutes after embolisation (right). (C) White light images (upper) and StO2 maps (lower) of the esophagus tissue (i) before the stomach removal, (ii) after the stomach removal, (iii) two minutes after the KCl injection, (iv) four minutes after the KCl injection and (v) twenty minutes after the KCl injection.
Figure 5.
StO2 maps obtained in human subject research.
(A) White light image by endoscopic observation in rectal adenocarcinoma (left). Line (L-R) corresponds to cross-section of pathological diagnosis. StO2 map visualized by laser endoscope system (middle: pseudocolor StO2 image; right: StO2 overlay image). (B) Cross-section appearance stained with H&E (upper) and HIF1 alpha antibody (lower) corresponding to the hypoxic area visualized with StO2 map. (C) Endoscopic images of a colorectal adenoma (upper) showing clear hypoxia: white light image (upper left), pseudocolor StO2 map (upper middle) and overlayed image (upper right). Another case of a colonic lesion (lower) consisting of an adenoma (red arrow) and a hyperplasia (blue arrow): white light image (lower left), pseudocolor StO2 map (lower middle) and overlayed image (lower right). Only the adenoma was detected as hypoxia. (D) Observed StO2 differences between neoplastic and non-neoplastic areas: For comparing pathology specimens and endoscope images, the line on the endoscopic image corresponding to the cross-section was determined. StO2 levels at neoplasic and non-neoplasic areas along this line were then calculated using this StO2 map.
Table 1.
Patients Characteristics.
Table 2.
Clinicopathological findings of the study with gastric cancer patients.