Computer-aided system for predicting the histology of colorectal tumors by using narrow-band imaging magnifying colonoscopy (with video).

BACKGROUND: Narrow-band imaging (NBI) classification of colorectal lesions is clinically useful in determining treatment options for colorectal tumors. There is a learning curve, however. Accurate NBI-based diagnosis requires training and experience. In addition, objective diagnosis is necessary. Thus, we developed a computerized system to automatically classify NBI magnifying colonoscopic images. OBJECTIVE: To evaluate the utility and limitations of our automated NBI classification system. DESIGN: Retrospective study. SETTING: Department of endoscopy, university hospital. MAIN OUTCOME MEASUREMENTS: Performance of our computer-based system for classification of NBI magnifying colonoscopy images in comparison to classification by two experienced endoscopists and to histologic findings. RESULTS: For the 371 colorectal lesions depicted on validation images, the computer-aided classification system yielded a detection accuracy of 97.8% (363/371); sensitivity and specificity of types B-C3 lesions for a diagnosis of neoplastic lesion were 97.8% (317/324) and 97.9% (46/47), respectively. Diagnostic concordance between the computer-aided classification system and the two experienced endoscopists was 98.7% (366/371), with no significant difference between methods. LIMITATIONS: Retrospective, single-center in this initial report. CONCLUSION: Our new computer-aided system is reliable for predicting the histology of colorectal tumors by using NBI magnifying colonoscopy.

Microvascular structure and perfusion imaging of colon cancer by means of contrast-enhanced ultrasonography.

OBJECTIVE: To determine whether contrast-enhanced ultrasonography (CEUS) can be used to discriminate between colon cancer and acute inflammation, and between mucinous and non-mucinous carcinoma. MATERIALS AND METHODS: CEUS (with perflubutane microbubbles) was performed in two study groups: colon cancer (n = 34) and acute inflammation (n = 14). For evaluation, the microvascular structure was classified as irregular or regular, and vessel diameter was classified as ≥2 mm or <2 mm. Tumor enhancement was classified as homogeneous, heterogeneous (obvious defect), or hypoenhancement. Moreover, the defect area was classified according to the presence or absence of vessels. Differences in imaging features between the two groups or between types of tumors were examined statistically. RESULTS: The vascular structure was irregular in 76.5% of colon cancers but only 28.6% of acute inflammations (P < 0.01). A significantly greater number of cancers contained vessels ≥2 mm (70.6% vs. 7.1%) (P < 0.001). Both abnormalities were found in 58.8% of colon cancers but in none of the acute inflammations. Enhancement patterns differed between tumor types, with mucin pools being readily identifiable. CONCLUSION: Differentiation between colon cancer and acute inflammation is possible with CEUS. Furthermore, prediction of mucinous vs. non-mucinous adenocarcinoma is possible.

Quantitative analysis and development of a computer-aided system for identification of regular pit patterns of colorectal lesions.

BACKGROUND: Because pit pattern classification of colorectal lesions is clinically useful in determining treatment options for colorectal tumors but requires extensive training, we developed a computerized system to automatically quantify and thus classify pit patterns depicted on magnifying endoscopy images. OBJECTIVE: To evaluate the utility and limitations of our automated pit pattern classification system. DESIGN: Retrospective study. SETTING: Department of endoscopy at a university hospital. MAIN OUTCOME MEASUREMENTS: Performance of our automated computer-based system for classification of pit patterns on magnifying endoscopic images in comparison to classification by diagnosis of the 134 regular pit pattern images by an endoscopist. RESULTS: For type I and II pit patterns, the results of discriminant analysis were in complete agreement with the endoscopic diagnoses. Type IIIl was diagnosed in 29 of 30 cases (96.7%) and type IV was diagnosed in 1 case. Twenty-nine of 30 cases (96.7%) were diagnosed as type IV pit pattern. The overall accuracy of our computerized recognition system was 132 of 134 (98.5%). CONCLUSIONS: Our system is best characterized as semiautomated but is a step toward the development of a fully automated system to assist in the diagnosis of colorectal lesions based on classification of pit patterns.

Comparison of detectability of small-bowel lesions between capsule endoscopy and double-balloon endoscopy for patients with suspected small-bowel disease.

BACKGROUND: There are many reports of the usefulness of capsule endoscopy (CE) or double-balloon endoscopy (DBE) for the detection of small-bowel disease. However, there are few reports that compared CE and DBE. OBJECTIVE: To determine whether CE or DBE better detects small-bowel lesions in patients with suspected small-bowel disease. DESIGN: A prospective single-center study. SETTING: Department of Endoscopy, Hiroshima University Hospital. MAIN OUTCOME MEASUREMENTS: Rates of CE and DBE detection of small-bowel lesions. PATIENTS: Seventy-six consecutive patients (47 men, 29 women; mean age 56.0 years) who underwent both CE and DBE. METHODS: CE was performed before DBE. DBEs were performed within 1 week, by both retrograde and antegrade approaches so that the entire small bowel could be examined, if possible. RESULTS: Small-bowel lesions were detected by CE in 42 patients (55.3%) and by DBE in 46 patients (60.5%). The difference was not significant (P = .45). Total enteroscopy was achieved by both examinations in 35 patients, and small-bowel lesions were detected by both examinations in 21 of the 35 patients (60.0%). Agreement between results of the 2 examinations was good (kappa = 0.76). LIMITATION: The main indication for examinations was not the same. CONCLUSIONS: CE and DBE are nearly equal in their ability to detect small-bowel lesions if the entire small bowel is examined.

Predicting the clinical response to cytapheresis in steroid-refractory or -dependent ulcerative colitis using contrast-enhanced ultrasonography.

OBJECTIVE: To evaluate the usefulness of transabdominal ultrasound (US), including contrast-enhanced ultrasonography (CEUS), in predicting the response to cytapheresis therapy in patients with steroid-refractory or -dependent ulcerative colitis (UC). MATERIAL AND METHODS: Between January 2005 and June 2008, 26 consecutive patients with steroid-refractory or -dependent UC were treated with granulocyte and monocyte adsorption apheresis (GCAP) or leukocytapheresis (LCAP) at our institute. The clinical activity of UC was evaluated by patients' C-reactive protein (CRP) levels and clinical activity index (CAI) scores. All patients were evaluated by grey-scale US, power Doppler US (PDUS), and CEUS. In CEUS, the color signal patterns were classified as 1 of 2 patterns. In pattern 1, color signals were partially detected in the bowel wall (excluding muscularis propria, the outer thin layer of the bowel wall), whereas in pattern 2, color signals were detected in the entire bowel wall (excluding muscularis propria). Differences between remission or clinical response (group R) and no response (group N) were ascertained for clinical features, clinical activities, and US findings. RESULTS: Differences between the two groups were not considered significant for the clinical features, clinical activities, and grey-scale US and PDUS findings. Using CEUS, 4 patients in group R showed pattern 2 (21%), while in group N, all patients showed this pattern, indicating a significant difference between the two groups (p<0.01). CONCLUSION: CEUS findings may be helpful in predicting the clinical response to cytapheresis for steroid-refractory or -dependent UC.

Optical biopsy of early gastroesophageal cancer by catheter-based reflectance-type laser-scanning confocal microscopy.

Magnified endoscopic observation of the gastrointestinal tract has become possible. However, such observation at the cellular level remains difficult. Laser-scanning confocal microscopy (LCM) is a novel, noninvasive optical imaging method that provides instant microscopic images of untreated tissue under endoscopy. We compare prototype catheter-based reflectance-type LCM images in vivo and histologic images of early gastroesophageal cancer to assess the usefulness of LCM in diagnosing such cancer. 20 sites in the esophagus and 40 sites in the stomach are examined by LCM under endoscopy prior to endoscopic or surgical resection. A prototype catheter LCM system, equipped with a semiconductor laser that oscillates at 685 nm and analyzes reflected light (Mauna Kea Technologies, Paris, France; Fujinon, Saitama, Japan) is used in vivo without fluorescent agent. In all normal esophageal mucosa and esophageal cancers, the nuclei are visualized. In nine of the ten normal esophageal mucosa, cell membranes are visualized, and in five of the ten esophageal cancers, cell membranes are visualized. In all normal gastric mucosa, nuclei and cell membranes are not visualized, but in ten of the 20 gastric cancers, nuclei are visualized. This novel method will aid in immediate diagnosis under endoscopy without the need for biopsy.

Computer-aided colorectal tumor classification in NBI endoscopy using local features.

An early detection of colorectal cancer through colorectal endoscopy is important and widely used in hospitals as a standard medical procedure. During colonoscopy, the lesions of colorectal tumors on the colon surface are visually inspected by a Narrow Band Imaging (NBI) zoom-videoendoscope. By using the visual appearance of colorectal tumors in endoscopic images, histological diagnosis is presumed based on classification schemes for NBI magnification findings. In this paper, we report on the performance of a recognition system for classifying NBI images of colorectal tumors into three types (A, B, and C3) based on the NBI magnification findings. To deal with the problem of computer-aided classification of NBI images, we explore a local feature-based recognition method, bag-of-visual-words (BoW), and provide extensive experiments on a variety of technical aspects. The proposed prototype system, used in the experiments, consists of a bag-of-visual-words representation of local features followed by Support Vector Machine (SVM) classifiers. A number of local features are extracted by using sampling schemes such as Difference-of-Gaussians and grid sampling. In addition, in this paper we propose a new combination of local features and sampling schemes. Extensive experiments with varying the parameters for each component are carried out, for the performance of the system is usually affected by those parameters, e.g. the sampling strategy for the local features, the representation of the local feature histograms, the kernel types of the SVM classifiers, the number of classes to be considered, etc. The recognition results are compared in terms of recognition rates, precision/recall, and F-measure for different numbers of visual words. The proposed system achieves a recognition rate of 96% for 10-fold cross validation on a real dataset of 908 NBI images collected during actual colonoscopy, and 93% for a separate test dataset.

Outcome of patients who have undergone total enteroscopy for obscure gastrointestinal bleeding.

AIM: To assess the diagnostic success and outcome among patients with obscure gastrointestinal bleeding who underwent total enteroscopy with double-balloon endoscopy. METHODS: Total enteroscopy was attempted in 156 patients between August 2003 and June 2008 at Hiroshima University Hospital and achieved in 75 (48.1%). It is assessed whether sources of bleeding were identified, treatment methods, complications, and 1-year outcomes (including re-bleeding) after treatment, and we compared re-bleeding rates among patients. RESULTS: The source of small bowel bleeding was identified in 36 (48.0%) of the 75 total enteroscopy patients; the source was outside the small bowel in 11 patients (14.7%) and not identified in 28 patients (37.3%). Sixty-one of the 75 patients were followed up for more than 1 year (27.2 ± 13.3 mo). Four (6.6%) of these patients showed signs of re-bleeding during the first year, but bleeding did not recur after treatment. Although statistical significance was not reached, a marked difference was found in the re-bleeding rate between patients in whom total enteroscopy findings were positive (8.6%, 3/35) and negative (3.8%, 1/26) (3/35 vs 1/26, P = 0.63). CONCLUSION: A good outcome can be expected for patients who undergo total enteroscopy and receive proper treatment for the source of bleeding in the small bowel.

Quantitative analysis of colorectal lesions observed on magnified endoscopy images.

BACKGROUND: Various surface mucosal pit patterns, as recognized by endoscopists, correlate with the histologic features of colorectal cancers. We investigated whether magnified endoscopy images of these pit patterns could be analyzed quantitatively and thus facilitate computer-aided diagnosis of colorectal lesions. METHODS: We applied both texture analysis and scale-invariant feature transform (SIFT) descriptors and discriminant analysis to magnified endoscopy images of 165 neoplastic colorectal lesions (pit patterns: type III(L)/IV, n = 44; type V(I)-mildly irregular, n = 36; type V(I)-severely irregular, n = 45; type V(N), n = 40) [histologic findings: tubular adenoma (TA), n = 56; carcinoma with intramucosal or even scant submucosal invasion (M/SM-s), n = 52, carcinoma with massive submucosal invasion (SM-m), n = 57]. We analyzed differences in pit pattern values and corresponding histologic values to determine whether the values were diagnostically meaningful. RESULTS: Gray-level difference matrix (GLDM) inverse difference moment and spatial gray-level dependence matrix (SGLDM) local homogeneity values differed significantly between type III(L)/IV and type V(N) pit patterns. Values differed significantly for each analyzed feature between type III(L)/IV and type V(I)-severely irregular patterns and were high but descending for type III(L)/IV, type V(I)-mildly irregular, and type V(I)-severely irregular pit patterns (in that order). Similarly, texture analysis yielded high but descending values for TA, M/SM-s, and SM-m (in that order). Furthermore, SIFT descriptors and discriminant analysis yielded differences that were superior to those obtained by texture analyses. CONCLUSIONS: Computer analysis of magnified endoscopy images for the diagnosis of colorectal lesions appears feasible. We anticipate further developments in the computer-aided diagnosis of pit patterns on magnified endoscopy images.