An automated machine vision system for the histological grading of cervical intraepithelial neoplasia (CIN)

The histological grading of cervical intraepithelial neoplasia (CIN) remains subjective, resulting in inter- and intra-observer variation and poor reproducibility in the grading of cervical lesions. This study has attempted to develop an objective grading system using automated machine vision. The architectural features of cervical squamous epithelium are quantitatively analysed using a combination of computerized digital image processing and Delaunay triangulation analysis; 230 images digitally captured from cases previously classified by a gynaecological pathologist included normal cervical squamous epithelium (n = 30), koilocytosis (n = 46), CIN 1 (n = 52), CIN 2 (n = 56), and CIN 3 (n=46). Intra- and inter-observer variation had kappa values of 0.502 and 0.415, respectively. A machine vision system was developed in KS400 macro programming language to segment and mark the centres of all nuclei within the epithelium. By object-oriented analysis of image components, the positional information of nuclei was used to construct a Delaunay triangulation mesh. Each mesh was analysed to compute triangle dimensions including the mean triangle area, the mean triangle edge length, and the number of triangles per unit area, giving an individual quantitative profile of measurements for each case. Discriminant analysis of the geometric data revealed the significant discriminatory variables from which a classification score was derived. The scoring system distinguished between normal and CIN 3 in 98.7% of cases and between koilocytosis and CIN 1 in 76.5% of cases, but only 62.3% of the CIN cases were classified into the correct group, with the CIN 2 group showing the highest rate of misclassification. Graphical plots of triangulation data demonstrated the continuum of morphological change from normal squamous epithelium to the highest grade of CIN, with overlapping of the groups originally defined by the pathologists. This study shows that automated location of nuclei in cervical biopsies using computerized image analysis is possible. Analysis of positional information enables quantitative evaluation of architectural features in CIN using Delaunay triangulation meshes, which is effective in the objective classification of CIN. This demonstrates the future potential of automated machine vision systems in diagnostic histopathology. Copyright (C) 2000 John Wiley and Sons, Ltd.

Assisted Diagnosis of Cervical Intraepithelial Neoplasia (CIN)

This paper introduces an automated computer- assisted system for the diagnosis of cervical intraepithelial neoplasia (CIN) using ultra-large cervical histological digital slides. The system contains two parts: the segmentation of squamous epithelium and the diagnosis of CIN. For the segmentation, to reduce processing time, a multiresolution method is developed. The squamous epithelium layer is first segmented at a low (2X) resolution. The boundaries are further fine tuned at a higher (20X) resolution. The block-based segmentation method uses robust texture feature vectors in combination with support vector machines (SVMs) to perform classification. Medical rules are finally applied. In testing, segmentation using 31 digital slides achieves 94.25% accuracy. For the diagnosis of CIN, changes in nuclei structure and morphology along lines perpendicular to the main axis of the squamous epithelium are quantified and classified. Using multi-category SVM, perpendicular lines are classified into Normal, CIN I, CIN II, and CIN III. The robustness of the system in term of regional diagnosis is measured against pathologists' diagnoses and inter-observer variability between two pathologists is considered. Initial results suggest that the system has potential as a tool both to assist in pathologists' diagnoses, and in training.

Cigarette smoking and high-risk HPV DNA as predisposing factors for high-grade cervical intraepithelial neoplasia (CIN) in young Brazilian women

Background. This cross-sectional study was designed to evaluate the role of cigarette smoking and high-risk HPV types as risk factors of CIN 2 and 3 in young, sexually active Brazilian women. Materials and method. A series of 100 consecutive women with abnormal Pap smears were recruited, subjected to colposcopy, punch biopsy, and questionnaire for their social, sexual and reproductive factors. Of these, 77 women between 20 and 35 years of age (median 26.5 years) with biopsy-confirmed CIN 1 or CIN 2 and 3, were enrolled in this study. Representative samples from the exocervix and endocervix were obtained for HPV testing with the Hybrid Capture HPV-DNA assay, including the probes for the oncogenic HPV types (16, 18, 31, 33, 35, 45, 51, 52 and 56). Results. The overall rate of CIN 2 and 3 was 23/77 (29.8%). The women with CIN 1, 2 and 3 did not differ from each other with regard to their age, race, schooling, marital status, life-time number of sexual partners, age at first intercourse, use of oral contraceptives, or parity. However, current cigarette smoking was strongly associated with CIN 2 and 3 (p < 0,001), and among smokers, the risk of high-grade CIN increased in parallel with the time of exposure (years of smoking) p = 0.07), HPV-DNA of the oncogenic types was detected in 43 (56%) women, the risk of being HPV DNA-positive was significantly higher in CIN 2 and 3 as compared with CIN 1 (p = 0.037). Importantly, the prevalence of high-risk HPV types was significantly higher in cigarette smokers than in non-smokers (p = 0.046). Conclusions. The results indicate that the severity of CIN lesions was clearly related to two fundamental risk factors: 1) high-risk HPV types, and 2) current cigarette smoking. These two risk factors were closely interrelated in that the high-risk HPV types were significantly more frequent in current smokers than in non-smokers, suggesting the possibility of a synergistic action between these two risk factors in cervical carcinogenesis.

Investigación y desarrollo en repositorios institucionales : aplicaciones y experiencias en universidades de la región bonaerense: Proyecto PICT-O CIN Región Bonaerense

Fil: Miguel, Sandra. Universidad Nacional de La Plata. Facultad de Humanidades y Ciencias de la Educación. Instituto de Investigaciones en Humanidades y Ciencias Sociales (UNLP-CONICET); Argentina.

Investigación y desarrollo en repositorios institucionales : aplicaciones y experiencias en universidades de la región bonaerense: Proyecto PICT-O CIN Región Bonaerense

Fil: Miguel, Sandra. Universidad Nacional de La Plata. Facultad de Humanidades y Ciencias de la Educación. Instituto de Investigaciones en Humanidades y Ciencias Sociales (UNLP-CONICET); Argentina.

Méthode structurelle pour le suivi automatique des artères coronaires en ciné-angiographie

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

La focalisation dans les romans, les ciné-romans et les films d'Alain Robbe-Grillet

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Méthode structurelle pour le suivi automatique des artères coronaires en ciné-angiographie

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

La focalisation dans les romans, les ciné-romans et les films d'Alain Robbe-Grillet

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Platelet-derived endothelial cell growth factor expression and angiogenesis in cervical intraepithelial neoplasia and squamous cell carcinoma of the cervix

Growth and metastatic spread of invasive carcinoma depends on angiogenesis, the formation of new blood vessels. Platelet-derived endothelial cell growth factor (PD-ECGF) is an angiogenic growth factor for a number of solid tumors, including lung, bladder, colorectal, and renal cell cancer. Cervical intraepithelial neoplasia (CIN) is the precursor to squamous cell cervical carcinoma (SCC). Mean vessel density (MVD) increases from normal cervical tissue, through low- and high-grade CIN to SCC. We evaluated PD-ECGF immunoreactivity and correlated its expression with MVD in normal, premalignant, and malignant cervical tissue. PD-ECGF expression was assessed visually within the epithelial tissues and scored on the extent and intensity of staining. MVD was calculated by counting the number of vessels positive for von Willebrand factor per unit area subtending normal or CIN epithelium or within tumor hotspots for SCC. Cytoplasmic and/or nuclear PD-ECGF immunoreactivity was seen in normal epithelium. PD-ECGF expression significantly increased with histologic grade from normal, through low- and high-grade CIN, to SCC (P < .02). A progressive significant increase in the microvessel density was also seen, ranging from a mean of 28 vessels for normal tissue to 57 for SCC (P < .0005). No correlation was found between PD-ECGF expression and MVD (P = .45). We conclude that PD-ECGF expression and MVD increase as the cervix transforms from a normal to a malignant phenotype. PD-ECGF is thymidine phosphorylase, a key enzyme in the activation of fluoropyrimidines, including 5-fluorouracil. Evaluation of PD-ECGF thymidine phosphorylase expression may be important in designing future chemotherapeutic trials in cervical cancer. Copyright (C) 2000 by W.B. Saunders Company.

High level of MT-MMP expression is associated with invasiveness of cervical cancer cells

MMP-2 (gelatinase A) has been associated with the invasive potential of many cancer cells both in vitro and in vivo. It is now becoming clear that the activation of this enzyme might be a key step in tumor invasion. This activation process has been shown to be a membrane-associated pathway inducible by various agents such as collagen type I, concanavalin A or TGF-β, but its physiological regulation is still largely unresolved. MT-MMP was recently discovered and described as a potential gelatinase-A activator. In the present study, we investigated the expression of MT-MMP (membrane-type metalloproteinase) in cervical cancer cells both in vitro and in vivo. Comparing several in vitro-transformed cervical cell lines, previously shown to display different invasive potentials, our results showed that the ability of cells to overexpress MT-MMP mRNA following ConA induction correlated with their ability to activate gelatinase A and with a highly invasive behavior. Moreover, using immunohistochemistry and in situ hybridization, we found a higher level of MT-MMP expression in invasive cervical carcinoma and lymphnode metastases compared to its expression in non-invasive CIN III lesions. Our in vivo observations also clearly demonstrated a cooperation between stromal and tumor cells for the production of MT-MMP. Taken together, our results clearly correlated high level MT-MMP expression with invasiveness, and thus suggested that MT-MMP might play a crucial role in cervical tumor invasion.

Vimentin expression in cervical carcinomas : association with invasive and migratory potential

Vimentin is an intermediate filament protein normally expressed in mesenchymal cells, but evidence is accumulating in the literature which suggests that the aberrant expression of vimentin in epithelial cancer cells might be related to local invasiveness and metastatic potential. Vimentin expression has previously been associated with invasive properties in an in vitro model consisting of a set of HPV-33-transformed cervical keratinocyte cell lines. In the present study, in order to emphasize those in vitro findings, the expression of vimentin has been investigated in cervical neoplasms of different grades, using immunohistochemistry. A clear association is reported between vimentin expression and metastatic progression, since vimentin was detected in all invasive carcinomas and lymph node metastases, but not in CIN III lesions. These in vivo results are compared with present and previous data obtained in vitro on cervical keratinocyte cell lines, where vimentin expression also correlated with in vitro invasiveness.