نوع مقاله : مقاله های پژوهشی
نویسندگان
1 دانشجوی کارشناسی ارشد، گروه بیوالکتریک و مهندسی پزشکی، دانشکدهی فناوریهای نوین علوم پزشکی و کمیتهی تحقیقات دانشجویی، دانشگاه علوم پزشکی اصفهان، اصفهان، ایران
2 دانشیار، گروه آسیبشناسی، دانشکدهی پزشکی، دانشگاه علوم پزشکی اصفهان، اصفهان، ایران
3 دانشیار، گروه بیوالکتریک و مهندسی پزشکی، دانشکدهی فناوریهای نوین علوم پزشکی، دانشگاه علوم پزشکی اصفهان، اصفهان، ایران
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Background: Vulvovaginal candidiasis (VVC) is a common clinical problem due to occurrence overgrowth of candida in genital system mucosa of females. The aim of this study was automatic diagnosis of vulvovaginal candidiasis via detection and extraction of candida fungus from microscopic images of Pap smear samples. We used image processing techniques to detect candida fungus.Methods: The sample space consisted of 200 microscopic images. Microscopic images were prepared from 49 Pap smear samples using Nikon1 V1 camera mounted on Nikon Eclipse 50i light microscope. For uniform illumination of the images, bottom-hat filtering was used. De-correlation stretching and linear contrast stretching were used for contrast enhancement. Different geometric features such as area, major axis, minor axis, eccentricity, perimeter, compactness, and decision tree classifier were used for extraction of mycelium and conidium of candida.Findings: The results of extraction of mycelium showed a specificity of 98.64% and a sensitivity of 96.88%. The corresponding values for conidium detection were 91.54% and 92.32%, respectively.Conclusion: According to our findings, this software would be helpful to pathologists in the diagnosis of vulvovaginal candidiasis in prevention of eyestrain. It could increase the accuracy of diagnosis, too.
کلیدواژهها [English]