Segmentation of cervical cell images using mean-shift filtering and morphological operators

Bergmeir, Christoph; Silvente, Miguel García; López-Cuervo, J. Esquivias; Benitez, Jose M.

doi:10.1117/12.845587

Cited by 17 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…[22] Early approaches to segmentation included geometric image analysis techniques such as mean-shift, median filtering, adaptive thresholding, Canny edge detection, edge detection by Riemannian dilatation, and Hough transform for finding candidate nuclei. [232425] Doudkine et al approached the segmentation problem by analyzing texture features from slides stained with quantitative stains for DNA. They used descriptive statistics of chromatin distribution, discrete texture features, ranges, Markovian, run length and fractal texture features for image classification.…”

Section: Discussionmentioning

confidence: 99%

Performance of a convolutional neural network in screening liquid based cervical cytology smears

Sanyal¹,

Barui²,

Deb

et al. 2019

J Cytol

View full text Add to dashboard Cite

Context: Cervical cancer is the second most common cancer in women. The liquid based cervical cytology (LBCC) is a useful tool of choice for screening cervical cancer. Aims: To train a convolutional neural network (CNN) to identify abnormal foci from LBCC smears. Settings and Design: We have chosen retrospective study design from archived smears of patients undergoing screening from cervical cancer by LBCC smears. Materials and Methods: 2816 images, each of 256 × 256 pixels, were prepared from microphotographs of these LBCC smears, which included 816 “abnormal” foci (low grade or high grade squamous intraepithelial lesion) and 2000 ‘normal’ foci (benign epithelial cells and reactive changes). The images were split into three sets, Training, Testing, and Evaluation. A convolutional neural network (CNN) was developed with the python programming language. The CNN was trained with the Training dataset; performance was assayed concurrently with the Testing dataset. Two CNN models were developed, after 20 and 10 epochs of training, respectively. The models were then run on the Evaluation dataset. Statistical Analysis Used: A contingency table was prepared from the original image labels and the labels predicted by the CNN. Results: Combined assessment of both models yielded a sensitivity of 95.63% in detecting abnormal foci, with 79.85% specificity. The negative predictive value was high (99.19%), suggesting potential utility in screening. False positives due to overlapping cells, neutrophils, and debris was the principal difficulty met during evaluation. Conclusions: The CNN shows promise as a screening tool; however, for its use in confirmatory diagnosis, further training with a more diverse dataset will be required.

show abstract

Section: Discussionmentioning

confidence: 99%

Performance of a convolutional neural network in screening liquid based cervical cytology smears

Sanyal¹,

Barui²,

Deb

et al. 2019

J Cytol

View full text Add to dashboard Cite

show abstract

“…With recent progress on image segmentation, the above models have also been applied to microscopic image segmentation, such as mean shift clustering for cytology nuclei segmentation [15], MRF-based active contour for glandular boundary extraction [156], and active contours for histology image segmentation [37,62]. Specifically for histology image segmentation, we have developed two other models: Gaussian mixture model based pixel labeling [67] and color clustering [66], in addition to our active contour method [68].…”

Section: Histology Image Processing and Analysismentioning

confidence: 99%

Histology image analysis for carcinoma detection and grading

Long

Antani

et al. 2012

Computer Methods and Programs in Biomedicine

318

184

View full text Add to dashboard Cite

This paper presents an overview of the image analysis techniques in the domain of histopathology, specifically, for the objective of automated carcinoma detection and classification. As in other biomedical imaging areas such as radiology, many computer assisted diagnosis (CAD) systems have been implemented to aid histopathologists and clinicians in cancer diagnosis and research, which have been attempted to significantly reduce the labor and subjectivity of traditional manual intervention with histology images. The task of automated histology image analysis is usually not simple due to the unique characteristics of histology imaging, including the variability in image preparation techniques, clinical interpretation protocols, and the complex structures and very large size of the images themselves. In this paper we discuss those characteristics, provide relevant background information about slide preparation and interpretation, and review the application of digital image processing techniques to the field of histology image analysis. In particular, emphasis is given to state-of-the-art image segmentation methods for feature extraction and disease classification. Four major carcinomas of cervix, prostate, breast, and lung are selected to illustrate the functions and capabilities of existing CAD systems.

show abstract

“…Aquí también los autores reconocen explícitamente que automatizar el proceso de segmentación de células cervicales sigue siendo un problema abierto debido a las complejidades de las estructuras en las células. En [10] se sigue insistiendo en la abrumadora dificultad en delimitar el núcleo dentro de la célula, debido a la amplia variabilidad de los núcleos entre células. Utilizan un procedimiento interactivo de determinación de regiones de interés sobre versiones de baja resolución y luego utilizan la máxima resolución para detectar las células dentro de esas regiones mediante filtros de mediana y detección de bordes.…”

Section: Introductionunclassified

Segmentación de imágenes de células cervicales y evaluación de características para detección de lesiones neoplásicas

Mejía

Rubiano

Alzate

2016

Ingeniare. Rev. chil. ing.

View full text Add to dashboard Cite

El cáncer cervical puede ser curado si se detecta y trata oportunamente, para ello la prueba Pap ha sido fundamental. En este contexto, una ayuda tecnológica puede reducir la naturaleza subjetiva del diagnóstico, pero existen dificultades en su construcción. Aquí abordamos dos de ellas: la identificación del citoplasma y el núcleo de cada célula, y la determinación de un conjunto de características relevantes para la detección de lesiones neoplásicas. En este artículo presentamos dos aportes. Primero se propone un método interactivo de segmentación basado en procesamiento morfológico multiespectral en el que las imperfecciones más engañosas de las imágenes son eliminadas con una interacción simple del analista. Segundo, se hace un análisis de la relevancia de algunas variables que caracterizan los tamaños relativos del núcleo y el citoplasma, sus formas, sus texturas y la rugosidad de sus bordes. El análisis se basa en las medidas de desempeño de un detector que utiliza extracción de características mediante análisis de componentes principales (PCA) y separación de células normales y lesionadas mediante una máquina de vectores de soporte (SVM). Encontramos que una mínima interacción con el médico permite obtener segmentaciones mucho más precisas y confiables. De otro lado, encontramos que las características más relevantes para detección de lesiones neoplásicas son los tamaños relativos del núcleo y del citoplasma y sus formas, mientras que otras características, como la textura y la rugosidad, son menos relevantes.

show abstract

Segmentation of cervical cell images using mean-shift filtering and morphological operators

Cited by 17 publications

References 0 publications

Performance of a convolutional neural network in screening liquid based cervical cytology smears

Performance of a convolutional neural network in screening liquid based cervical cytology smears

Histology image analysis for carcinoma detection and grading

Segmentación de imágenes de células cervicales y evaluación de características para detección de lesiones neoplásicas

Contact Info

Product

Resources

About