Abstract
The aim of the present study was to design an adaptable pattern recognition (PR) system to discriminate low- from high-grade squamous intraepithelial lesions (LSIL and HSIL, respectively) of the cervix using microscopy images of hematoxylin and eosin (H&E)-stained biopsy material from two different medical centers. Clinical material comprised H&E-stained biopsies of 66 patients diagnosed with LSIL (34 cases) or HSIL (32 cases). Regions of interest were selected from each patient’s digitized microscopy images. Seventy-seven features were generated, regarding the texture, morphology and spatial distribution of nuclei. The probabilistic neural network (PNN) classifier, the exhaustive search feature selection method, the leave-one-out (LOO) and the bootstrap validation methods were used to design the PR system and to assess its precision. Optimal PR system design and evaluation were made feasible by the employment of graphics processing unit (GPU) and Compute Unified Device Architecture (CUDA) technologies. The accuracy of the PR-system was 93% and 88.6% when using the LOO and bootstrap validation methods, respectively. The proposed PR system for discriminating LSIL from HSIL of the cervix was designed to operate in a clinical environment, having the capability of being redesigned when new verified cases are added to its repository and when data from other medical centers are included, following similar biopsy material preparation procedures.
Acknowledgments
The authors would like to thank the Departments of Pathology of the University Hospital of Patras and IASO Thessalias medical center, for supplying and assisting in the peer evaluation of the H&E-stained histology specimens.
Author statement
Research funding: None declared.
Conflict of interest: None declared.
Informed consent: Informed consent is not applicable.
Ethical approval: The study was conducted in accordance with the guidelines of the Declaration of Helsinki and of the Ethics Committee of the University of Patras, Greece.
References
[1] Kyrgiou M, Shafi M. Colposcopy and cervical intraepithelial neoplasia. Obstet Gynaecol Reprod Med 2014;24:204–14.10.1016/j.ogrm.2014.05.002Search in Google Scholar
[2] De S, Stanley RJ, Lu C, Long R, Antani S, Thoma G, et al. A fusion-based approach for uterine cervical cancer histology image classification. Comput Med Imaging Graph 2013;37:475–87.10.1016/j.compmedimag.2013.08.001Search in Google Scholar
[3] Sedivy R, Windischberger C, Svozil K, Moser E, Breitenecker G. Fractal analysis: an objective method for identifying atypical nuclei in dysplastic lesions of the cervix uteri. Gynecol Oncol 1999;75:78–83.10.1006/gyno.1999.5516Search in Google Scholar
[4] Price GJ, McCluggage WG, Morrison MM, McClean G, Venkatraman L, Diamond J, et al. Computerized diagnostic decision support system for the classification of preinvasive cervical squamous lesions. Hum Pathol 2003;34:1193–203.10.1016/S0046-8177(03)00421-0Search in Google Scholar
[5] McCluggage WG, Walsh MY, Thornton CM, Hamilton PW, Date A, Caughley LM, et al. Inter- and intra-observer variation in the histopathological reporting of cervical squamous intraepithelial lesions using a modified Bethesda grading system. Br J Obstet Gynaecol 1998;105:206–10.10.1111/j.1471-0528.1998.tb10054.xSearch in Google Scholar
[6] Ismail SM, Colclough AB, Dinnen JS, Eakins D, Evans DM, Gradwell E, et al. Reporting cervical intra-epithelial neoplasia (CIN): intra- and interpathologist variation and factors associated with disagreement. Histopathology 1990;16:371–6.10.1111/j.1365-2559.1990.tb01141.xSearch in Google Scholar
[7] Keenan SJ, Diamond J, McCluggage WG, Bharucha H, Thompson D, Bartels PH, et al. An automated machine vision system for the histological grading of cervical intraepithelial neoplasia (CIN). J Pathol 2000;192:351–62.10.1002/1096-9896(2000)9999:9999<::AID-PATH708>3.0.CO;2-ISearch in Google Scholar
[8] Wang Y, Crookes D, Eldin OS, Wang S, Hamilton P, Diamond J. Assisted diagnosis of cervical intraepithelial neoplasia (CIN). IEEE J Sel Topics Signal Process 2009;3:112–21.10.1109/JSTSP.2008.2011157Search in Google Scholar
[9] Miranda GHB, Soares EG, Barrera J, Felipe JC, editors. Method to support diagnosis of cervical intraepithelial neoplasia (CIN) based on structural analysis of histological images. 25th International Symposium on Computer-Based Medical Systems 2012.10.1109/CBMS.2012.6266297Search in Google Scholar
[10] Rahmadwati R, Naghdy G, Ros M, Todd C. Computer aided decision support system for cervical cancer classification. In: SPIE, Applications of Digital Image Processing XXXV; 2012;8499:1–13.10.1117/12.931284Search in Google Scholar
[11] Guo P, Banerjee K, Joe Stanley R, Long R, Antani S, Thoma G, et al. Nuclei-based features for uterine cervical cancer histology image analysis with fusion-based classification. IEEE J Biomed Health Inform 2016;20:1595–607.10.1109/JBHI.2015.2483318Search in Google Scholar PubMed PubMed Central
[12] Guillaud M, Cox D, Adler-Storthz K, Malpica A, Staerkel G,Matisic J, et al. Exploratory analysis of quantitative histopathology of cervical intraepithelial neoplasia: objectivity, reproducibility, malignancy-associated changes, and human papillomavirus. Cytometry A 2004;60:81–9.10.1002/cyto.a.20034Search in Google Scholar PubMed
[13] Guillaud M, Cox D, Malpica A, Staerkel G, Matisic J, Van Niekirk D, et al. Quantitative histopathological analysis of cervical intra-epithelial neoplasia sections: methodological issues. Cell Oncol 2004;26:31–43.10.1155/2004/238769Search in Google Scholar PubMed PubMed Central
[14] Huang P, Chan Y, Chan P, Chen Y, Chen R, Huang Y. Quantitative assessment of pap smear cells by PC-based cytopathologic image analysis system and support vector machine. LNCS 2008;4901:192–9.10.1007/978-3-540-77413-6_25Search in Google Scholar
[15] Chen YF, Huang PC, Lin KC, Wang LE, Cheng CC, Chen TP, et al. Semi-automatic segmentation and classification of Pap smear cells. IEEE J Biomed Health Inform 2014;18:94–108.10.1109/JBHI.2013.2250984Search in Google Scholar PubMed
[16] Song D, Kim E, Huang X, Patruno J, Muñoz-Avila H, Heflin J, et al. Multimodal entity coreference for cervical dysplasia diagnosis. IEEE Trans Med Imaging 2015;34:229–45.10.1109/TMI.2014.2352311Search in Google Scholar PubMed
[17] Xu T, Zhang H, Xin C, Kim E, Long LR, Xue Z, et al. Multi-feature based benchmark for cervical dysplasia classification evaluation. Pattern Recognit 2017;63:468–75.10.1016/j.patcog.2016.09.027Search in Google Scholar PubMed PubMed Central
[18] Kyrgiou M, Pouliakis A, Panayiotides JG, Margari N, Bountris P, Valasoulis G. Personalised management of women with cervical abnormalities using a clinical decision support scoring system. Gynecol Oncol 2016;141:29–35.10.1016/j.ygyno.2015.12.032Search in Google Scholar PubMed
[19] Koeneman MM, van Lint FHM, van Kuijk SMJ, Smits LJM, Kooreman LFS, Kruitwagen RFPM, et al. A prediction model for spontaneous regression of cervical intraepithelial neoplasia grade 2, based on simple clinical parameters. Hum Pathol 2017;59:62–9.10.1016/j.humpath.2016.09.012Search in Google Scholar PubMed
[20] Wang HK, Duffy AA, Broker TR, Chow LT. Robust production and passaging of infectious HPV in squamous epithelium of primary human keratinocytes. Genes Dev 2009;23:181–94.10.1101/gad.1735109Search in Google Scholar PubMed PubMed Central
[21] Ambroise C, McLachlan GJ. Selection bias in gene extraction on the basis of microarray gene-expression data. Proc Natl Acad Sci USA 2002;99:6562–6.10.1073/pnas.102102699Search in Google Scholar
[22] Loukas CG, Wilson GD, Vojnovic B, Linney A. An image analysis-based approach for automated counting of cancer cell nuclei in tissue sections. Cytometry A 2003;55:30–42.10.1002/cyto.a.10060Search in Google Scholar
[23] Haralick R, Shanmugam K, Dinstein I. Textural features for image classification. IEEE Trans Syst Man Cybern 1973;3: 610–21.10.1109/TSMC.1973.4309314Search in Google Scholar
[24] Galloway MM. Texture analysis using gray-level run lengths. Comput Graph Imag Process 1975;4:172–9.10.1016/S0146-664X(75)80008-6Search in Google Scholar
[25] Theodoridis S, Koutroumbas K. Pattern Recognition. 2nd ed. San Diego: Elsevier; 2003.Search in Google Scholar
[26] Tamura H, Mori S, Yamawaki T. Textural features corresponding to visual perception. IEEE Trans Syst Man Cybern B Cybern 1978;8:460–73.10.1109/TSMC.1978.4309999Search in Google Scholar
[27] Niedermayer I, Kolles H, Feiden W. Histologic and morphometric grading of gliomas. A comparative survival analysis. Anal Quant Cytol Histol 1997;19:301–10.Search in Google Scholar
[28] Sidiropoulos K, Glotsos D, Kostopoulos S, Ravazoula P, Kalatzis I, Cavouras D, et al. Real time decision support system for diagnosis of rare cancers, trained in parallel, on a graphics processing unit. Comput Biol Med 2012;42:376–86.10.1016/j.compbiomed.2011.12.004Search in Google Scholar
[29] Specht D. Probabilistic neural networks. Neural Networks 1990;3:109–18.10.1016/0893-6080(90)90049-QSearch in Google Scholar
[30] Glotsos D, Spyridonos P, Petalas P, Cavouras D, Ravazoula P, Dadioti PA, et al. Computer-based malignancy grading of astrocytomas employing a support vector machine classifier, the WHO grading system and the regular hematoxylin-eosin diagnostic staining procedure. Anal Quant Cytol Histol 2004;26:77–83.Search in Google Scholar
[31] Glotsos D, Kalatzis L, Spyridonos P, Kostopoulos S, Daskalakis A, Athanasiadis E, et al. Improving accuracy in astrocytomas grading by integrating a robust least squares mapping driven support vector machine classifier into a two level grade classification scheme. Comput Meth Prog Bio 2008;90:251–61.10.1016/j.cmpb.2008.01.006Search in Google Scholar PubMed
[32] Ninos K, Kostopoulos S, Sidiropoulos K, Kalatzis I, Glotsos D, Athanasiadis E, et al. Computer-based image analysis system designed to differentiate between low-grade and high-grade laryngeal cancer cases. Anal Quant Cytol Histol 2013;35: 261–72.Search in Google Scholar
[33] Kostopoulos S, Konstandinou C, Sidiropoulos K, Ravazoula P, Kalatzis I, Asvestas P, et al. Assessing the performance of four different categories of histological criteria in brain tumours grading by means of a computer-aided diagnosis image analysis system. J Microsc 2015;260:37–46.10.1111/jmi.12264Search in Google Scholar PubMed
[34] Ninos K, Kostopoulos S, Kalatzis I, Sidiropoulos K, Ravazoula P, Sakellaropoulos G, et al. Microscopy image analysis of p63 immunohistochemically stained laryngeal cancer lesions for predicting patient 5-year survival. Eur Arch Otorhinolaryngol 2016;273:159–68.10.1007/s00405-015-3747-xSearch in Google Scholar PubMed
[35] Konstandinou C, Glotsos D, Kostopoulos S, Kalatzis I, Ravazoula P, Michail G. Multifeature quantification of nuclear properties from images of H&E-stained biopsy material for investigating changes in nuclear structure with advancing CIN grade. J Healthc Eng 2018;2018:6358189.10.1155/2018/6358189Search in Google Scholar PubMed PubMed Central
©2019 Walter de Gruyter GmbH, Berlin/Boston
