Descriptores MAC en la detección temprana del cáncer cervical usando técnicas de procesamiento de imágenes

Resumen

Cervical cancer is a global public health problem. Much progress has been made to improve its early detection. Despite Brazil’s efforts to reduce mortality indicators from this disease, they are still not sufficient compared to progress in other countries. The objective of this study is to present descriptors of the characteristics used in Malignity Associated Changes (MAC) in cervical cells for early detection of cervical cancer using Image Processing, Computer-Assisted.  As a result, the 400 most used MAC descriptors were identified and their accuracy was calculated to be 86,94%. A naive Bayes net classifier was used. The potential of computational tracking technologies that can assist in the early identification of cervical cell anomalies and improve CCU tracking capabilities in Brazil and Latin American countries was evidenced.