A Unified Model-Based Image Analysis Framework for Automated Detection of Precancerous Lesions in Digitized Uterine Cervix Images

Women’s cancers remain a major challenge for many health systems. Between 1991 and 2017, the death rate for all major cancers fell continuously in the United States, excluding uterine cervix and uterine corpus cancers. Together with HPV (Human Papillomavirus) testing and cytology, colposcopy has played a central role in cervical cancer screening. This medical procedure allows physicians to view the cervix at a magnification of up to 10%. This paper presents an automated colposcopy image analysis framework for the classification of precancerous and cancerous lesions of the uterine cervix. This framework is based on an ensemble of MobileNetV2 networks. Our experimental results show that this method achieves accuracies of 83.33% and 91.66% on the four-class and binary classification tasks, respectively. These results are promising for the future use of automatic classification methods based on deep learning as tools to support medical doctors.

Download Full-text

An Automated Image Analysis Framework for Model-Based Feature Detection in Sparse Data

10.18130/v3d247 ◽

2014 ◽

Author(s):

Andrea Vaccari

Keyword(s):

Image Analysis ◽

Feature Detection ◽

Sparse Data ◽

Automated Image Analysis ◽

Analysis Framework ◽

Model Based

Download Full-text

Quantitative Model-Based Image Analysis of NuMa Distribution Links Nuclear Organization with Cell Phenotype

Microscopy and Microanalysis ◽

10.1017/s1431927600028968 ◽

2001 ◽

Vol 7 (S2) ◽

pp. 578-579

Author(s):

David W. Knowles ◽

Sophie A. Lelièvre ◽

Carlos Ortiz de Solόrzano ◽

Stephen J. Lockett ◽

Mina J. Bissell ◽

...

Keyword(s):

Image Analysis ◽

Mammary Epithelial Cells ◽

Nuclear Organization ◽

Critical Role ◽

Quantitative Model ◽

Mammary Epithelial ◽

Cell Phenotype ◽

Proliferating Cells ◽

Mitotic Apparatus ◽

Model Based

The extracellular matrix (ECM) plays a critical role in directing cell behaviour and morphogenesis by regulating gene expression and nuclear organization. Using non-malignant (S1) human mammary epithelial cells (HMECs), it was previously shown that ECM-induced morphogenesis is accompanied by the redistribution of nuclear mitotic apparatus (NuMA) protein from a diffuse pattern in proliferating cells, to a multi-focal pattern as HMECs growth arrested and completed morphogenesis . A process taking 10 to 14 days.To further investigate the link between NuMA distribution and the growth stage of HMECs, we have investigated the distribution of NuMA in non-malignant S1 cells and their malignant, T4, counter-part using a novel model-based image analysis technique. This technique, based on a multi-scale Gaussian blur analysis (Figure 1), quantifies the size of punctate features in an image. Cells were cultured in the presence and absence of a reconstituted basement membrane (rBM) and imaged in 3D using confocal microscopy, for fluorescently labeled monoclonal antibodies to NuMA (fαNuMA) and fluorescently labeled total DNA.

Download Full-text

Multicentre Evaluation of Hepika Test Clinical Accuracy in Diagnosing HPV-Induced Cancer and Precancerous Lesions of the Uterine Cervix

Diagnostics ◽

10.3390/diagnostics11040619 ◽

2021 ◽

Vol 11 (4) ◽

pp. 619

Author(s):

Daniela Gustinucci ◽

Lucia Ciccocioppo ◽

Luigi Coppola ◽

Giovanni Negri ◽

Gianfranco Zannoni ◽

...

Keyword(s):

Uterine Cervix ◽

Protein Complexes ◽

Precancerous Lesions ◽

Intraepithelial Neoplasia ◽

Population Based ◽

Enzyme Linked Immunosorbent Assay ◽

Lower Sensitivity ◽

Adenocarcinoma In Situ ◽

Clinical Accuracy ◽

Grade 3

Objective: To evaluate the clinical accuracy of Hepika test to identify cancer/precancerous lesions of the uterine cervix. Materials and Methods: A multicentre retrospective study was carried out in 2018 and included 330 liquid-based cytology samples from three Italian centres of women aged 25–64 who had been tested for the human papillomavirus (HPV) and whose histology or follow-up outcome was known. Hepika is an enzyme-linked immunosorbent assay (ELISA) targeting the protein complexes E6#p53 and E7#pRb. After excluding samples without sufficient residual material, the clinical accuracy of Hepika test was evaluated in 274 samples: adenocarcinoma (ADC) (4), squamous cell carcinoma (SCC) (7), adenocarcinoma in situ (AIS) (1), cervical intraepithelial neoplasia (CIN) grade 3 (60), CIN2 (51), CIN1 (34), and negative histology (117). Association, sensitivity, and specificity for carcinoma, CIN3+ and CIN2+ are reported. Results: Positive Hepika test was associated with a high probability of carcinoma (odds ratio (DOR) = 33.68, 95% confidence interval (CI) 7.0–163.1); sensitivity was 81.8%, specificity, 88.2%. A positive Hepika test showed a weaker association with CIN3+ lesions (DOR = 3.5; 95% CI 1.75–6.99) and lower sensitivity (27.8%). Conclusion: The Hepika test was found to be an accurate biomarker for HPV-induced cervical carcinoma. Population-based prospective studies are needed to confirm the clinical usefulness of the Hepika test in the differential diagnosis of HPV-induced invasive lesions.

Download Full-text