| Title |
Reikšmingų uždegiminių žarnyno ligų diagnostinių požymių identifikavimas histologiniuose žarnyno audinių vaizduose / |
| Translation of Title |
Identification of diagnostic features of intestine inflammatory diseases in histological tissue images. |
| Authors |
Kiškis, Andrius |
| Full Text |
|
| Pages |
53 |
| Keywords [eng] |
ulcerative colitis ; Crohn’s disease ; histology |
| Abstract [eng] |
Ulcerative colitis (UC) and Crohn’s disease (CD) are two principal diseases of chronic inflammatory bowel disease (IBD). For visual evaluation of inflammatory processes according to bowel histological samples, pathology experts use qualitative estimates of infringement – presence or absence of certain signs. However, this evaluation is not always right because of the “human error”. There is a great possibility that the same expert evaluating the same sample at different time period will get different results. However, a reliable and fast automatic methodology for diagnostics of inflammatory bowel diseases has not been still developed. For automatic evaluation of histological samples, inflammatory cells forming during inflammation should be considered. Hence, the aim of this work is to create a method for detection of inflammatory cells in histological sample images of mice colon tissue. In order to achieve the goal these tasks were carried out. Image preprocessing was done for registered pictures. Uneven illumination was eliminated and optimal color space transformation was performed by means of principle component analysis images. By the help of mathematical morphology methods probable inflammatory cell areas were detected in the image. The elaborated method then performs feature extraction procedures and discriminant analysis identified the most important ones. Obtained results demonstrated the potential use of created method for histology specialist assistance. |
| Dissertation Institution |
Kauno technologijos universitetas. |
| Type |
Master thesis |
| Language |
Lithuanian |
| Publication date |
2015 |
