Pattern recognition on 2D cervical cytological digital images for early detection of cervix cancer

Abstract

To date, cancer of the uterine cervix is still a leading cause of cancer-related deaths in women in the world. Papanicolau smear test is a well-known screening method of detecting abnormalities in the uterine cervix cells. In Indonesia, Pap smear test is mostly still done conventionally. Due to the small number of skilled and experienced cytologists, the screening procedure becomes time consuming and highly prone to human errors. Coping with these issues, an automated recognition system is developed to enable automatic identification of anomaly in the cervix cells. Recognition of patterns inside a cervix cell is based on the cell morphological features, in terms of size, shape, and color. Therefore, three parameters are employed, i.e. N/C ratio, wavelet approximation coefficients, and color intensity. Based on thorough observation upon the selected parameters, it can be recognized that the cancerous cells follow certain patterns and highly distinguishable from the normal cells.

Sistema Experto Probabilístico basado en Redes Bayesianas para la predicción del cáncer de cuello uterino

“…En el año 2009 se presentaron dos casos de estudio relacionados al cáncer de cuello uterino, uno de ellos en Indonesia, donde el Departamento de Ingeniería Biomédica de la Universidad Swiss German [6] presentó un sistema para la detección temprana de cáncer de cuello uterino utilizando reconocimiento de patrones sobre imágenes citológicas de la zona cervical en 2D, con el fin de lidiar con el problema de inconsistencia e imprecisión del diagnóstico del test del Papanicolaou. Por otro lado, Onísko, Druzdzel y Marshall [7] implementaron un modelo basado en Redes Bayesianas Dinámicas al cual dieron el nombre de PCCSM (Pittsburgh Cervical Cancer Screening Model), el cual permite evaluar el riesgo de cáncer cervical invasivo en el transcurso del tiempo, con la finalidad de obtener una serie de resultados por paciente en diferente etapas cronológicas, lo cual ayudaría a los médicos a realizar un seguimiento individual a cada uno de ellos para tomar decisiones más certeras al momento de programar las evaluaciones médicas respectivas.…”

Intelligent Screening Systems for Cervical Cancer

“…Several techniques are applied to extract the size and shape features: thresholding technique [ 60 , 62 , 77 , 81 ]; clustering technique [ 70 , 73 ]; fuzzy technique [ 69 ]; wavelet technique [ 82 , 83 ]; statistic techniques [ 13 , 22 , 78 , 84 ]. …”

“…At cellular-level, the size and shape features in cytology images are extracted using thresholding [ 62 ], clustering [ 70 , 73 ], fuzzy [ 69 ], and wavelet techniques [ 83 ]. In the FISH images, the features are extracted using thresholding [ 60 , 77 , 81 ].…”

“…Cytology image has a relatively darker color intensity composition than normal cells. The distinguishable patterns can be analyzed using the corresponding image's color intensity histogram [ 83 ]. Meanwhile, another feature to differentiate the abnormality of cervix using FISH image is the average intensity of each colored spot [ 60 ].…”