RSS-Feed abonnieren
DOI: 10.1055/s-2004-814519
Computerized Diagnosis of Helicobacter pylori Infection and Associated Gastric Inflammation from Endoscopic Images by Refined Feature Selection Using a Neural Network
Publikationsverlauf
Submitted 28 February 2003
Accepted after Revision 19 February 2004
Publikationsdatum:
09. Juli 2004 (online)
![](https://www.thieme-connect.de/media/endoscopy/200407/lookinside/thumbnails/10.1055-s-2004-814519-1.jpg)
Background and Study Aim: We investigated whether analysis of endoscopic images using a refined feature selection with neural network (RFSNN) technique could predict Helicobacter pylori-related gastric histological features.
Patients and Methods: A total of 104 dyspeptic patients were prospectively enrolled for panendoscopy and gastric biopsy for histological evaluation using the updated Sydney system. The endoscopic images of each patient were analyzed to obtain 84 image parameters. The significant image parameters from 30 randomly selected patients (15 with and 15 without H. pylori infection) associated with histological features were used to develop the RFSNN model. This was then used to test the sensitivity and specificity of the image parameters obtained from the remaining 74 patients for the prediction of the presence of H. pylori infection and related histological features.
Results: The RFSNN technique had a sensitivity of 85.4 % and a specificity of 90.9 % for the detection of H. pylori infection. Moreover, RFSNN was highly accurate (> 80 %) in predicting the presence of gastric atrophy, intestinal metaplasia and the severity of H. pylori-related gastric inflammation.
Conclusions: RFSNN is an effective computerized technique for assessing the presence of H. pylori infection and related gastric inflammation and precancerous lesions. By using RFSNN to analyze endoscopic images, a comprehensive evaluation of the stomach may be done, thus avoiding the need for invasive but localized biopsy sampling for histological examination.
References
- 1 Parsonnet J, Friedman G F, Vandersteen D P. et al . H. pylori infection and the risk of gastric carcinoma. N Engl J Med. 1991; 325 1127-1131
- 2 Walsh J H, Peterson W L. The treatment of Helicobacter pylori infection in the management of peptic ulcer disease. N Engl J Med. 1995; 333 984-991
- 3 McColl K EL, El-Nujumi A, Murray L. et al . The Helicobacter pylori breath test: a surrogate marker for peptic ulcer disease in dyspeptic patients. Gut. 1997; 40 302-306
- 4 Dixon M F, Genta R M, Yardley J H, Correa P. Classification and grading of gastritis: the updated Sydney system. International Workshop on the Histology of Gastritis, Houston 1994. Am J Surg Pathol. 1996; 20 1161-1181
- 5 Yang S B, Sheu B S, Su I J. et al . Clinical application of gastric histology to monitor treatment of dual therapy in H. pylori eradication. Dig Dis Sci. 1997; 42 1835-1840
- 6 Sheu B S, Yang S B, Su I J. et al . Bacterial density of H. pylori predicts the success of triple therapy in bleeding duodenal ulcer. Gastrointest Endosc. 1996; 44 683-688
- 7 Sheu B S, Wu J J, Yang H B. et al . One week proton pump inhibitor-based triple therapy is effective in eradicating residual H. pylori after failed dual therapy. J Formos Med Assoc. 1998; 97 266-270
- 8 Sheu B S, Yang H B, Wang Y L. et al . Pretreatment gastric histology is predictive to the symptomatic response after H. pylori eradication of patients with non-ulcer dyspepsia. Dig Dis Sci. 2001; 46 2700-2707
- 9 Robert M, Haralick K S, Dinstein I. Textural features for image classification. IEEE T Syst Man CY. 1973; 3 610-621
- 10 Unser M. Sum and difference histograms for texture classification. IEEE T Pattern Anal. 1986; 8 118-125
- 11 Haykin S. Neural networks: a comprehensive foundation. 2nd edn. New Jersey; Prentice-Hall 1999
- 12 Rumelhart D E, Hinton G E, Williams R J. Learning representations of back-propagation errors. Nature. 1986; 323 533-536
- 13 Parker D B. Optimal algorithms for adaptive networks: second order back propagation, second order direct propagation, and second order Hebbian learning. IEEE 1st International Conference of Neural Networks. San Diego, California; 1987
- 14 Correa P, Miller M J. Helicobacter pylori and gastric atrophy - cancer paradoxes. J Natl Cancer Inst. 1995; 87 1731-1732
- 15 Tabata H, Fuchigami T, Kobayashi H. et al . Helicobacter pylori and mucosal atrophy in patients with gastric cancer. Dig Dis Sci. 1999; 44 2027-2034
- 16 Jang T J, Kim D I, Shin Y M. et al . Carcinogenesis - p16INK4A promoter hypermethylation of non-tumorous tissue adjacent to gastric cancer is correlated with glandular atrophy and chronic inflammation. Intl J Cancer. 2001; 93 629-634
- 17 Kashiwagi H. Ulcer and gastritis. Endoscopy. 2003; 35 9-14
- 18 Yagi K, Nakamura A, Sekine A. Comparison between magnifying endoscopy and histological, culture and urease test findings from the gastric mucosa of corpus. Endoscopy. 2002; 34 376-381
B.-S. Sheu, M. D
Department of Internal Medicine · National Cheng Kung University Hospital
138 Sheng-Li Road · Tainan 70428 · Taiwan
Fax: +886-6-2370941
eMail: sheubs@mail.ncku.edu.tw