Celiac disease (CD) is a common multisystemic immune-mediated disease. CD is associated with considerable morbidity and mortality. For the diagnosis of CD, biopsy of the small intestine is still the gold standard and is usually recommended in most of the children as CD requires a lifelong commitment to a strict GFD. CD usually arises in the small intestine leading to mucosal alterations classified into different stages according to severity. Typical mucosal alterations found in CD patients include flattening and/or broadening of the intestinal villi, elongation of the crypts and an increase of the number of intraepithelial lymphocytes.
However, the histological examination of biopsy specimens from the duodenum suffers from a significant degree of intra- and inter-observer variability. Therefore, there is a need of observer independent diagnostic methods overcoming this limitation of histopathology. Computer-aided detection of villous atrophy is highly reliable based on computer assisted pattern analysis methods performed on endoscopic still images taken in the so-called modified immersion technique (MIT). MIT allows detailed scanning of the mucosal surface for villous structures. Technically, water is rapidly instilled into the duodenal lumen after evacuation of air by suction through the endoscope. Villi, if present, straighten up in water and appear as tiny finger-like structures.
If automated analysis can detect villous atrophy not only in endoscopic still images taken in MIT but in endoscopic video clips as well, in such cases, a small bowel biopsy could entirely be avoided. The avoidance of biopsy not only increases the safety of upper endoscopy, but also contributes to significant cost savings in such cases, because there is no need for the time and resource intensive processing of biopsy specimens followed by histological examination. Another advantage is quality improvement rendered possible by means of telemedicine, for example, by obtaining a second opinion after endoscopic video clips have been transmitted to institutions that have video analysis software at their disposal allowing for an observer independent objective diagnostic evaluation.
A further limitation of the current gold standard for the diagnosis of celiac disease is due to the possibly patchy distribution of intestinal mucosa areas affected by CD in the midst of normal mucosa. If, unfortunately, biopsies are taken only from areas of healthy mucosa within the duodenum the proper diagnosis of CD will be missed due to sampling error. A diagnostic tool based on computer-aided pattern analysis of endoscopic video clips could indicate areas that are damaged by celiac disease thus targeting biopsy areas, if needed.
The main question we will try to assess is, if such software can be developed and easily integrated into the routine of pediatric endoscopy. Furthermore, we will examine if the narrow band imaging technique, an optical/digital chromoendoscopy, can further improve the reliability of automated analysis. Various feature extraction and classification strategies will be applied for automated differentiation between presence and absence of villous atrophy in the endoscopic video clips.
Gadermayr M, Kogler H, Karla M, Merhof D, Uhl A, Vécsei A. Computer-aided texture analysis combined with experts’ knowledge: Improving endoscopic celiac disease diagnosis. World J Gastroenterol 22: 7124-7134, 2016
Tringali A, Thomson M, Dumonceau J-M, Tavares M, Tabbers MM, Furlano R, Spaander M, Hassan C, Tzvinikos C, Ijsselstijn H, Viala J, Dall’Oglio L, Benninga M, Orel R, Vandenplas Y, Keil R, Romano C, Brownstone E, Hlava S, Gerner P, Dolak W, Landi R, Huber WD, Everett S, Vecsei A, Aabakken L, Amil-Dias J, Zambelli A. Pediatric GI Endoscopy: Eur Soc of GI Endoscopy (ESGE) and Eur Soc of Ped Gastroenterol Hepatol and Nutrition (ESPGHAN) Guideline Executive summary. Endoscopy 2016. In press.
Kosulin K, Geiger E, Vécsei A, Huber W-D, Rauch M, Brenner E, Wrba F, Hammer K, Innerhofer A, Pötschger U, Lawitschka A, Matthes-Leodolter S, Fritsch G, Lion T. Persistence and reactivation of human adenoviruses in the gastrointestinal tract. Clin Microbiol Infect 22: 381.e1-8, 2016
Hegenbart S, Uhl A, Vécsei A. Survey on computer aided decision support for diagnosis of celiac disease. Computers in Biology and Medicine 65: 348-358, 2015
Vécsei E, Steinwendner S, Kogler H, Innerhofer A, Hammer K, Haas OA, Amann G, Chott A, Vogelsang H, Schoenlechner R, Huf W, Vécsei A. Follow-up of pediatric celiac disease: value of antibodies in predicting mucosal healing, a prospective cohort study. BMC Gastroenterol 14: 28, 2014
Vécsei A, Amann G, Hegenbart S, Kwitt R, Liedlgruber M, Uhl A. Automated Marsh-like classification of celiac disease in children using local texture operators. Comput Biol Med 41: 313-325, 2011
Vécsei A, Innerhofer A, Graf U, Binder C, Gizci H, Hammer K, Bruckdorfer A, Hirschl AM, Makristathis A. Helicobacter pylori eradication rates in children upon susceptibility testing based on noninvasive stool polymerase chain reaction versus gastric tissue culture. J Pediatr Gastroenterol Nutr 53: 65-70, 2011
Vécsei A, Kipet A, Innerhofer A, Graf U, Binder C, Gizci H, Hammer K, Bruckdorfer A, Huber WD, Hirschl AM, Makristathis A. Time trends of Helicobacter pylori resistance to antibiotics in children living in Vienna, Austria. Helicobacter 15: 214-220, 2010
Vécsei A, Innerhofer A, Binder C, Gizci H, Hammer K, Bruckdorfer A, Riedl S, Gadner H, Hirschl AM, Makristathis A. Stool polymerase chain reaction for Helicobacter pylori detection and Clarithromycin susceptibility testing in children. Clin Gastroenterol Hepatol 8: 309-312, 2010