Abstract
<jats:p>Cervical cancer is the fourth most common cancer in women worldwide and the third most prevalent type among women in Brazil. The mortality rate in Brazil, adjusted to the world population, was 5.33 deaths per 100,000 women in the latest triennium (2019-2022). Cervical cancer screening uses as a first line the traditional Papanicolaou test, visual inspection with acetic acid and Lugol's iodine, liquid-based cytology (LBC) and, in some cases, molecular tests for oncogenic HPV subtypes. Although advances have been made over the years, conventional screening procedures still suffer from diagnostic subjectivity, leading to false-positive and false-negative results. The present study aims to compare the main cytomorphological features that distinguish low-grade (LSIL) and high-grade (HSIL) squamous intraepithelial lesions, by means of images captured through optical microscopy and through the Pickcells automated system using the Pickcells Path scanner. To this end, five samples of each intraepithelial lesion, LSIL and HSIL, were selected. The criteria analysed, based on the Bethesda system, were the nuclear-cytoplasmic ratio, hyperchromasia, nuclear membrane contour and pleomorphism. The results obtained by the automated capture method were superior to the manual method, as they better highlighted the cytomorphological patterns characteristic of the lesions. Therefore, computer-assisted or automated cytology emerges as a perspective in the diagnosis of cervical carcinoma. The growing interest in the development of software and automated systems for cervical cancer screening aims to optimise analysis time, avoid ambiguity, eliminate false negatives and minimise the mortality rate.</jats:p>