INTESTINE Vol.10 No.3(2-3)

Theme Development of diagnosis of magnifying endoscopy
Title Examination of surface microstructure of large intestinal tumors
Publish Date 2006/05
Author Nobuyuki Obinata Division of Gastroenterology, Akita Red Cross Hospital / Division of Gastroenterology and Hepatology, St. Marianna University School of Medicine Hospital
Author Hiro-o Yamano Division of Gastroenterology, Akita Red Cross Hospital
Author Hiro-o Matsushita Division of Gastroenterology, Akita Red Cross Hospital / Division of Surgery, Health Insurance Hitoyoshi General Hospital
Author Kentarou Satoh Division of Gastroenterology, Akita Red Cross Hospital
Author Kenjirou Yoshikawa Division of Gastroenterology, Akita Red Cross Hospital
Author Ayako Kotake Division of Gastroenterology, Akita Red Cross Hospital
Author Kazuhiko Fujita Division of Gastroenterology and Hepatology, St. Marianna University School of Medicine Hospital
Author Fumio Itoh Division of Gastroenterology and Hepatology, St. Marianna University School of Medicine Hospital
[ Summary ] Purpose : To objectively examine the surface microstructure with large intestinal tumors by image analysis.
Subject : Lesions [e. g., laterally spreading non-granular tumor (LST NG)] that can be examined with sufficiently clear images, obtained using magnifying endoscopy, stereoscopic microscopy and light microscopy of histopathological specimens, were examined, along with large intestinal tumors, in patients treated at the Division of Gastroenterology, Akita Red Cross Hospital.
Method : On the basis of the histopathological features observed with a light microscopy image, detected lesions were divided into three groups : severe atypical adenoma (hereafter, sev group), m and sm slight invasive carcinoma (m / sm-S group) and sm deep invasive carcinoma (sm-M group). The endoscopic and stereoscopic microscopy images of these lesions were analyzed using image-processing software (MacScope Ver. 2). The morphological characteristic values (the number of pits, size of lesion, complexity value and areas of pit per 1 mm2) as well as the statistical characteristic value [dye contrast value (staining intensity) reflecting depth of pit] were examined.
Results : The morphological characteristic values are as follows : (1) Numbers of pits ; 68.0 (sev group), 40.6 (m / sm-S group), and 14.5 (sm-M group) (2) Sizes of lesions ; 0.44 x 10-2 mm2 (sev group), 0.48 x 10-2 mm2 (m / sm-S group), and 0.83 x 10-2 mm2 (sm-M group) (3) Complexity values ; 23.1 (sev group), 38.3 (m / sm-S group), and 46.8 (sm-M group) (4) Areas of pit per 1 mm2 ; 0.30 x 10-2 mm2 (sev group), 0.15 x 10-2 mm2 (m / sm-S group), and 0.11 x 10-2 mm2 (sm-M group). The statistical characteristic values are as follows : dye contrast values (depth of pit) ; 12.4 (sev group), 14.0 (m / sm-S group), and 8.59 (sm-M group). The pit depth of the sm-M group was deeper than those of the other two groups.
Statistically significant differences were confirmed between the sm-M group and the sev and m / sm-S groups in all the results obtained in this study.
Conclusions : The surface microstructure of large intestinal tumors was examined using image-analyzing software. Based on the morphological characteristic values and statistical characteristic values, an objective diagnosis is possible using a magnifying endoscope. We plan to report on the results of further examinations in the future.
back