Article

A standardised protocol for texture feature analysis of endoscopic images in gynaecological cancer.

Department of Computer Science, University of Cyprus (UCY), Nicosia, Cyprus.
BioMedical Engineering OnLine (Impact Factor: 1.61). 02/2007; 6:44. DOI: 10.1186/1475-925X-6-44
Source: DOAJ

ABSTRACT In the development of tissue classification methods, classifiers rely on significant differences between texture features extracted from normal and abnormal regions. Yet, significant differences can arise due to variations in the image acquisition method. For endoscopic imaging of the endometrium, we propose a standardized image acquisition protocol to eliminate significant statistical differences due to variations in: (i) the distance from the tissue (panoramic vs close up), (ii) difference in viewing angles and (iii) color correction.
We investigate texture feature variability for a variety of targets encountered in clinical endoscopy. All images were captured at clinically optimum illumination and focus using 720 x 576 pixels and 24 bits color for: (i) a variety of testing targets from a color palette with a known color distribution, (ii) different viewing angles, (iv) two different distances from a calf endometrial and from a chicken cavity. Also, human images from the endometrium were captured and analysed. For texture feature analysis, three different sets were considered: (i) Statistical Features (SF), (ii) Spatial Gray Level Dependence Matrices (SGLDM), and (iii) Gray Level Difference Statistics (GLDS). All images were gamma corrected and the extracted texture feature values were compared against the texture feature values extracted from the uncorrected images. Statistical tests were applied to compare images from different viewing conditions so as to determine any significant differences.
For the proposed acquisition procedure, results indicate that there is no significant difference in texture features between the panoramic and close up views and between angles. For a calibrated target image, gamma correction provided an acquired image that was a significantly better approximation to the original target image. In turn, this implies that the texture features extracted from the corrected images provided for better approximations to the original images. Within the proposed protocol, for human ROIs, we have found that there is a large number of texture features that showed significant differences between normal and abnormal endometrium.
This study provides a standardized protocol for avoiding any significant texture feature differences that may arise due to variability in the acquisition procedure or the lack of color correction. After applying the protocol, we have found that significant differences in texture features will only be due to the fact that the features were extracted from different types of tissue (normal vs abnormal).

0 Bookmarks
 · 
105 Views
  • Source
  • [Show abstract] [Hide abstract]
    ABSTRACT: The aims of this study were to compare two different methods of quantifying the colour of the luminal surface of the equine endometrium and to relate the results to histopathological evidence of inflammation and fibrosis. The mucosal surfaces of 17 equine uteri obtained from an abattoir were assessed using a spectrophotometer and by computer-assisted analysis of photographs. Values were converted into L(*)a(*)b(*) colour space. Although there was significant correlation between the two methods of quantification, variations in 'brightness', 'red' and 'yellow' values were noted. Within a given uterus, measurements using the spectrophotometer did not differ significantly. Using photographic analysis, brightness differed between horns, although no differences in chromaticity were found. Histopathological classification of changes within endometria corresponded to measured differences in colour. Extensive fibrosis was associated with increased brightness and decreased chromaticity using both methods. Inflammation correlated with reduced chromaticity, when measured by spectrophotometry, and with reduced brightness and yellow values, when assessed photographically. For this technique to gain wider acceptance as a diagnostic tool, e.g. for the endoscopic evaluation of uterine mucosae in vivo, standardised illumination techniques will be required so that colours can be compared and interpreted accurately.
    The Veterinary Journal 05/2013; · 2.42 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVES/HYPOTHESIS: Visualization of the vocal folds is essential when reaching a primary diagnosis of laryngeal disease. However, the examination is subjective and highly dependent on the experience of the treating physician. The present study is the development of objective tools for the diagnosis of laryngeal malignancy based on laryngeal texture analysis. STUDY DESIGN: Texture analysis using gray-level co-occurrence matrix (GLCM) in vocal fold images of 198 patients. METHODS: Vocal-fold images were subjected to texture analysis using gray-level co-occurrence matrix (GLCM)-based parameters, which were computed by a novel digital image-processing program. Patients were divided into two groups: those with benign-looking lesions and those with malignant-looking lesions. Textural irregularities were compared using GLCM-based parameters. The relationship between the texture-analysis parameters and the diagnosis was then statistically evaluated. RESULTS: Texture irregularity was negatively correlated with energy and the inverse difference moment (IDM) and positively correlated with entropy, variance, contrast, dissimilarity, and mean values. All of the GLCM-based parameters evaluated differed significantly according to the degree of differentiation of the benign- or malignant-looking lesion (P < 0.001). Entropy had a sensitivity of 82.9% and a specificity of 82.2% at a cutoff value of 5.94; for variance, the sensitivity was 82.9% and the specificity was 84.5% at a cutoff value of 167. CONCLUSION: GLCM-based texture analysis of vocal-fold lesions, especially in association with a differential diagnosis of benign and malignant-looking diseases, contributes to achieving an objective image-based analysis of vocal-fold lesions. In addition, this approach can be used to create algorithms permitting a reproducible classification of laryngeal pathologies. LEVEL OF EVIDENCE: NA. Laryngoscope, 2013.
    The Laryngoscope 05/2013; · 1.98 Impact Factor

Full-text (6 Sources)

Download
52 Downloads
Available from
May 30, 2014