Fractal characterization of chromatin appearance for diagnosis in breast cytology

Abstract

This study explores the use of fractal analysis in the numerical description of chromatin appearance in breast cytology. Images of nuclei from fine‐needle aspiration biopsies of the breast are characterized in terms of their Minkowski and spectral fractal dimensions, for 19 patients with benign epithelial cell lesions and 22 with invasive ductal carcinomas. Chromatin appearance in breast epithelial cell nuclear images is demonstrated to be fractal, suggesting that the three‐dimensional chromatin structure in these cells also has fractal properties. A statistically significant difference between the mean spectral dimensions of the benign and malignant cases is demonstrated. The two fractal dimensions are very weakly correlated. A statistically significant difference between the benign and malignant cases in lacunarity, a fractal property characterizing the size of holes or gaps in a texture, is found over a wide range of scales. These differences are particularly pronounced at the smallest and largest scales, corresponding respectively to fine‐scale texture, indicating whether chromatin is clumped or fine, and to large‐scale structures like nucleoli. Logistic regression and artificial neural network classification models are developed to classify unknown cases on the basis of fractal measures of chromatin texture. Using leave‐one‐out cross‐validation, the best logistic regression classifier correctly diagnoses 95·1 per cent of the cases. The best neural network model can correctly classify all of the cases, but it is unclear whether this is due to overtraining. Fractal dimensions and lacunarity are useful tools for the quantitative characterization of chromatin appearance, and can potentially be incorporated into image analysis devices to assure the quality and reproducibility of diagnosis by breast fine‐needle aspiration biopsy. © 1998 John Wiley & Sons, Ltd.

Identifying a risk profile for thyroid cancer

“…Pathologists have proposed the fractal dimension of nuclear chromatin as a prognostic factor in some tumors, such as lymphoblastic leukemia (91), myelodysplastic syndromes (92) and laryngeal carcinoma (93). This is a promising method that may estimate patients' outcome early on a cytological fine-needle aspirate, since it has also been used for fine-needle aspirates from breast cytology (94). Recent data suggest that this technique may be applied to thyroid and possibly be useful in predicting aggressiveness (95).…”

Techniques of Data Mining In Healthcare: A Review

“…This was one of the reasons which encouraged NN as the most widely used classification algorithm in various biomedicine and healthcare fields [44,45,46]. For example, NN has been widely used as the algorithm supporting the diagnosis of diseases including cancers [47][48][49][50][51] and predict outcomes [52][53][54]. In NN, basic elements are neurons or nodes.…”

Learning the Topological Properties of Brain Tumors

“…6 As a result, we set the values of the parameters as follows: the grid size is 4, the node threshold is 0.50, the link exponent is 4.4, and the number of hidden units is 12.…”

“…Another subset of these studies uses fractals that describe the similarity levels of different structures found in a tissue image over a range of scales [6], [9]. These studies use the fractal dimensions as their features and use the k-nearest neighborhood algorithm [9], neural networks, and logistic regression [6] as their classifiers.…”

“…These studies use the fractal dimensions as their features and use the k-nearest neighborhood algorithm [9], neural networks, and logistic regression [6] as their classifiers. Finally, the orientational features are extracted by making use of Gabor filters that respond to contrast edges and line-like features of a specific orientation [24].…”