Niki Margari

An Intelligent Clinical Decision Support System for Patient-Specific Predictions to Improve Cervical Intraepithelial Neoplasia Detection

Nowadays, there are molecular biology techniques providing information related to cervical cancer and its cause: the human Papillomavirus (HPV), including DNA microarrays identifying HPV subtypes, mRNA techniques such as nucleic acid based amplification or flow cytometry identifying E6/E7 oncogenes, and immunocytochemistry techniques such as overexpression of p16. Each one of these techniques has its own performance, limitations and advantages, thus a combinatorial approach via computational intelligence methods could exploit the benefits of each method and produce more accurate results. In this article we propose a clinical decision support system (CDSS), composed by artificial neural networks, intelligently combining the results of classic and ancillary techniques for diagnostic accuracy improvement. We evaluated this method on 740 cases with complete series of cytological assessment, molecular tests, and colposcopy examination. The CDSS demonstrated high sensitivity (89.4%), high specificity (97.1%), high positive predictive value (89.4%), and high negative predictive value (97.1%), for detecting cervical intraepithelial neoplasia grade 2 or worse (CIN2+). In comparison to the tests involved in this study and their combinations, the CDSS produced the most balanced results in terms of sensitivity, specificity, PPV, and NPV. The proposed system may reduce the referral rate for colposcopy and guide personalised management and therapeutic interventions.

Performance Evaluation of Manual and Automated (MagNA Pure) Nucleic Acid Isolation in HPV Detection and Genotyping Using Roche Linear Array HPV Test

Nucleic acids of human papillomavirus (HPV) isolated by manual extraction method (AmpliLute) and automated MagNA pure system were compared and evaluated with cytohistological findings in 253 women. The concordance level between AmpliLute and MagNA was very good 93.3% (κ = 0.864, P < .0001). Overall HPVpositivity detected by AmpliLute was 57.3% (30.4% as single and 27% as multiple infections) in contrast to MagNA 54.5% (32% and 23%, resp.). Discrepant results observed in 25 cases: 11 MagNA(−)/AmpliLute(+), 10 of which had positive histology; 5 MagNA(+)/AmpliLute(−) with negative histology; 8 MagNA(+)/AmpliLute(+): in 7 of which AmpliLute detected extra HPV genotypes and 1 MagNA(invalid)/AmpliLute(+) with positive histology. Both methods performed well when compared against cytological (area under curve (AUC) of AmpliLute 0.712 versus 0.672 of MagNA) and histological diagnoses (AUC of AmpliLute 0.935 versus 0.877 of MagNA), with AmpliLute showing a slightly predominance over MagNA. However, higher sensitivities, specificities, and positive/negative predictive values were obtained by AmpliLute.

Using classification and regression trees, liquid-based cytology and nuclear morphometry for the discrimination of endometrial lesions

‘The objective of this study is to investigate the potential of classification and regression trees (CARTs) in discriminating benign from malignant endometrial nuclei and lesions. The study was performed on 222 histologically confirmed liquid based cytological smears, specifically: 117 benign cases, 62 malignant cases and 43 hyperplasias with or without atypia. About 100 nuclei were measured from each case using an image analysis system; in total, we collected 22783 nuclei. The nuclei from 50% of the cases (the training set) were used to construct a CART model that was used for knowledge extraction. The nuclei from the remaining 50% of cases (test set) were used to evaluate the stability and performance of the CART on unknown data. Based on the results of the CART for nuclei classification, we propose two classification methods to discriminate benign from malignant cases. The CART model had an overall accuracy for the classification of endometrial nuclei equal to 85%, specificity 90.68%, and sensitivity 72.05%. Both methods for case classification had similar performance: overall accuracy in the range 94–95%, specificity 95%, and sensitivity 91–94%. The results of the proposed system outperform the standard cytological diagnosis of endometrial lesions. This study highlights interesting diagnostic features of endometrial nuclear morphology and provides a new classification approach for endometrial nuclei and cases. The proposed method can be a useful tool for the everyday practice of the cytological laboratory. Diagn. Cytopathol. 2014;42:582–591.

Artificial Neural Networks as Decision Support Tools in Cytopathology: Past, Present, and Future

Objective This study aims to analyze the role of artificial neural networks (ANNs) in cytopathology. More specifically, it aims to highlight the importance of employing ANNs in existing and future applications and in identifying unexplored or poorly explored research topics. Study Design A systematic search was conducted in scientific databases for articles related to cytopathology and ANNs with respect to anatomical places of the human body where cytopathology is performed. For each anatomic system/organ, the major outcomes described in the scientific literature are presented and the most important aspects are highlighted. Results The vast majority of ANN applications are related to cervical cytopathology, specifically for the ANN-based, semiautomated commercial diagnostic system PAPNET. For cervical cytopathology, there is a plethora of studies relevant to the diagnostic accuracy; in addition, there are also efforts evaluating cost-effectiveness and applications on primary, secondary, or hybrid screening. For the rest of the anatomical sites, such as the gastrointestinal system, thyroid gland, urinary tract, and breast, there are significantly less efforts relevant to the application of ANNs. Additionally, there are still anatomical systems for which ANNs have never been applied on their cytological material. Conclusions Cytopathology is an ideal discipline to apply ANNs. In general, diagnosis is performed by experts via the light microscope. However, this approach introduces subjectivity, because this is not a universal and objective measurement process. This has resulted in the existence of a gray zone between normal and pathological cases. From the analysis of related articles, it is obvious that there is a need to perform more thorough analyses, using extensive number of cases and particularly for the nonexplored organs. Efforts to apply such systems within the laboratory test environment are required for their future uptake.

Molecular epidemiology of HPV infection using a clinical array methodology in 2952 women in Greece

The molecular epidemiology of human papillomavirus (HPV) infection in a sample of Greek women (n = 2952, mean age 42.2 ± 13.3 years) was examined. HPV prevalence was 50.7% (95% confidence interval, 48.8-52.6). The most frequent HPV types were HPV 53, 51 and 66 (10.2%, 9.4% and 9.3%, respectively). HPV positivity was associated with age, age of sexual debut, number of sexual partners and duration of sexual relationship, while marriage or multiparity protected against infection (all p <0.001). Follow-up of this cohort will assist in predicting the effect of vaccination with the new HPV vaccines on future screening with HPV-based tests for cervical cancer.

Personalised management of women with cervical abnormalities using a clinical decision support scoring system

OBJECTIVES To develop a clinical decision support scoring system (DSSS) based on artificial neural networks (ANN) for personalised management of women with cervical abnormalities. METHODS We recruited women with cervical abnormalities and healthy controls that attended for opportunistic screening between 2006 and 2014 in 3 University Hospitals. We prospectively collected detailed patient characteristics, the colposcopic impression and performed a series of biomarkers using a liquid-based cytology sample. These included HPV DNA typing, E6&E7 mRNA by NASBA or flow cytometry and p16INK4a immunostaining. We used ANNs to combine the cytology and biomarker results and develop a clinical DSSS with the aim to improve the diagnostic accuracy of tests and quantify the individuals risk for different histological diagnoses. We used histology as the gold standard. RESULTS We analysed data from 2267 women that had complete or partial dataset of clinical and molecular data during their initial or followup visits (N=3565). Accuracy parameters (sensitivity, specificity, positive and negative predictive values) were assessed for the cytological result and/or HPV status and for the DSSS. The ANN predicted with higher accuracy the chances of high-grade (CIN2+), low grade (HPV/CIN1) and normal histology than cytology with or without HPV test. The sensitivity for prediction of CIN2 or worse was 93.0%, specificity 99.2% with high positive (93.3%) and negative (99.2%) predictive values. CONCLUSIONS The DSSS based on an ANN of multilayer perceptron (MLP) type, can predict with the highest accuracy the histological diagnosis in women with abnormalities at cytology when compared with the use of tests alone. A user-friendly software based on this technology could be used to guide clinician decision making towards a more personalised care.

Use of flow cytometry as a quality control device for liquid‐based cervical cytology specimens

Cervical cancer is the second most common cancer in women worldwide comprising a major concern of public health. Liquid‐based cytology provides significantly more effective detection of low‐grade intraepithelial neoplasia or more severe lesions, without loss of diagnostic specificity and reduces the number of cases classified as unsatisfactory. The objective of the study is the evaluation of flow cytometry as a rapid tool for quality control of the liquid specimen adequacy for the purpose of precise cytological diagnosis in detecting cervical abnormalities.

Image analysis and multi-layer perceptron artificial neural networks for the discrimination between benign and malignant endometrial lesions

This study aims to investigate the efficacy of an Artificial Neural Network based on Multi‐Layer Perceptron (ANN–MPL) to discriminate between benign and malignant endometrial nuclei and lesions in cytological specimens.

A reporting system for endometrial cytology: Cytomorphologic criteria—Implied risk of malignancy

There have been various attempts to assess endometrial lesions on cytological material obtained via direct endometrial sampling. The majority of efforts focus on the description of cytological criteria that lead to classification systems resembling histological reporting formats. These systems have low reproducibility, especially in cases of atypical hyperplasia and well differentiated carcinomas. Moreover, they are not linked to the implied risk of malignancy.

Assessment of contralateral mammary gland dose in the treatment of breast cancer using accelerated hypofractionated radiotherapy

AIM To measure the dose distribution, related to the treatment planning calculations, in the contralateral mammary gland of breast cancer patients treated with accelerated hypofractionated 3-dimensional conformal radiotherapy. METHODS Thirty-four prospectively selected female patients with right breast cancer (pN0, negative surgical margins) were treated with breast-conserving surgery. A total dose of 42.5 Gy (2.66 Gy/fraction) was prescribed; it was requested that planning target volumes be covered by the 95% isodose line. The contralateral mammary gland was defined on CT simulation. The dose received was evaluated by dose volume histograms. RESULTS The measured contralateral breast doses were: (1) Dose maximum: 290-448 cGy [Equivalent (Eq) 337-522 cGy]; (2) Mean dose: 45-70 cGy (Eq 524-815 cGy); and (3) Median dose: 29-47 cGy (337-547 cGy) for total primary breast dose of 42.5 Gy in 16 equal fractions. The spearman rho correlation showed statistical significance between the contralateral breast volume and maximum dose (P = 0.0292), as well as mean dose (P = 0.0025) and median dose (P = 0.046) to the breast. CONCLUSION Minimizing the dose to the contralateral breast has to be one of the priorities of the radiation oncologist when using short schedules because of the radiosensitivity of this organ at risk. Further study is necessary to assess the long-term clinical impact of this schedule.