Evaluation of Image Analysis in the Diagnosis of Cytologic Cervicovaginal

  The image analysis methods are auxiliary methods of pathological diagnosis, which can increase the accuracy and speed of evaluation of the samples.

This cross-sectional study was performed on 50 samples of cervicovaginal smear cytology from Pathology Department of Ghaem Hospital, Mashhad, Iran. Firstly, samples were divided into four diagnostic groups, including reactive epithelial changes, normal, low-grade squamous intraepithelial lesion (LSIL), and high-grade squamous intraepithelial lesion (HSIL) using Bethesda method. Afterwards, the images of cytology slides were prepared using an Olympus microscope imaging system and Olympus DP3 image. The next step included the analysis of factors, such as the area and the perimeter, gray value of cells.

The mean values of cell area were statistically different among the normal, LSIL, HSIL groups (P>0.01). The mean values of the circular index for three groups were not statistically different (P=0.151). The mean scores of gray value were significantly different between the normal and HSIL groups in terms (P=0.00), meaning that intensity of the color was higher in the nucleus of the HSIL group. This means that cells appear darker in the HSIL group (P>0.01).

Discussion & The obtained results of the study indicated that an image analysis could extract useful information from digital images related to cytology of cervical smear. The findings of the four diagnostic groups, namely reactive epithelial changes, normal, LSIL, HSIL were in line with those of pathologists’ eye evaluation