CLASSIFICATION METHODS FOR THE ANALYSIS OF SEGMENTED OBJECTS ON FLUORESCENT IMAGES OF CANCER CELLS
Article Sidebar
Main Article Content
Abstract
The methods of classification to analyze the multi-сhannel fluorescent images of breast cancer were studied. Each object is described by 13 features, where 11 features are geometry characteristics and 2 features corresponds to color characteristics. The methods were studied onb the standardized and not-standardized data. The cross validation was used. The considered methods are linear and quadratic discriminant analysis, Naive Bayes, Support Vector Machines, Decision Trees, Random Forest, Neural network models. The most sufficient result were received for the Random Forest methods, where the accuracy of the classification is 0,97, when all features are used. If only color features are exploited, the accuracy is 0,96, and finally it is 0,92 for form features. The same results received the linear discriminant analysis, where the accuracy based on all features is 0,97, the accuracy received by color features is 0,96. Which is the same as for random forest classification. However for form features it is only 0,90. The most unsufficient results are obtained for Multi-layer Perceptron.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
Программный пакет CELLDATAMINER для анализа люминесцентных изображений раковых кле-ток / Е.В. Лисица [и др.] // Информатика. – 2015. – № 4. – Стр. 73–84.
Третьяк, И.Ю. Клинико-морфологическая характеристика пациентов с отечноинфильтративной формой рака молочной железы / И.Ю. Третьяк, Ю.Е. Демидчик, С.А. Костюк // Актуальные вопросы ди-агностики и лечения злокачественных новообразований : материалы респ. науч.-практ. конф., посвященной 40-летию кафедры онкологии БГМУ, Минск, 21 нояб. 2014 / Белорус. гос. мед. ун-т ; под общ. ред. проф. А.В. Прохорова. – Минск, 2014. –С.102–106.
Paulin, F. and Santhakumaran, A. Extracting rules from feed forward neural networks for diagnosing breast cancer / F. Paulin, A. Santhakumaran // Artificial Intelligent Systems and Machine Learning. – 2009. – V. 1, № 4 – P. 143–146.
Global and Local Structure Preservation for Feature Selection / X. Liu [et al.] // Neural Networks and Learning Systems, IEEE Transactions. – 2013. – V. 99 – P. 1.
Hota, H. Diagnosis of breast cancer using intelligent techniques / H. Hota // Int J Emerg Sci Eng (IJESE). –2013. –V. 1, № 3 –P. 45–53.
Reif, M. Efficient feature size reduction via predictive forward selection / M. Reif, F. Shafait // Pattern Recognition. – 2014. – V. 47, № 4 – P. 1664–1673.
Cancer worldwide // World cancer report 2014 / B. Stewart, C. P. Wild. ‒ WHO, 2016. ‒ P. 16–81.
Gayathri, B. Breast cancer diagnosis using machine learning algorithms –ASurvey / B. Gayathri, C. Sumathi, T. Santhanam // International Journal of Distributed and Parallel Systems. – 2013. – 4(3). – P. 105.
Digital quantitative measurements of gene expression / V. Mikkilineni [et al.] // Biotechnology and bioengi-neering. ‒2004. ‒ № 86, V. 2. ‒ P. 117‒124.
Spatial quantitative analysis of fluorescently labeled nuclear structures: problems, methods, pitfalls / O. Ronneberger [et al.] // Chromosome Res. ‒ 2008. ‒ № 16, V. 3. ‒ P. 523‒562.
Алгоритм автоматической сегментации границ ядер раковых клеток на трехканальных люминесцентных изображениях / Лисица Е.В. [и др.] // Журнал прикладной спектроскопии. – 2015. – № 82, Т. 4. – С. 598–607.
Stabenfeldt, S.E. Current trends in biomarker discovery and analysis tools for traumatic brain injury / S.E. Stabenfeldt, B.I. Martinez // Journal of Biological Engineering. ‒ 2019. ‒ № 13, V. 16. ‒ P. 1–12.
Интеллектуальный анализ данных : пособие / Н.Н. Яцков. – Минск : БГУ, 2014. – 151 с.
MaGIC: a machine learning tool set and web application for monoallelic gene inference from chromatin / S. Vinogradova [et al.] // BMC Bioinformatics. – 2019. – № 20, V. 1. – P. 106–111.
Neumann, U. EFS: an ensemble feature selection tool implemented as R-package and web-application / U. Neumann, N. Genze, D. Heider // BioData Mining. – 2017. – № 10, V. 21.
Отбор характеристик распределения интенсивности в цветовых каналах на люминесцентных изображениях раковых клеток / Е. В. Лисица [и др.] // Журнал прикладной спектроскопии. – 2019. – Т. 86, № 3 – С. 394–400.
Отбор информативных геометрических признаков ядер клеток на люминесцентных изображениях раковых клеток / Е. В. Лисица [и др.] // Информатика. – 2019. – Т. 16, № 2 – С. 16–26.
Machine Learning-based Analysis of Rectal Cancer MRI Radiomics for Prediction of Metachronous Liver Metastasis / M. Liang [et al.] // Acad Radiol. – 2019. [Epub ahead of print].
Constructing Prediction Model of Lung Cancer Recurrence Risk Using Gene Function Clustering and Machine Learning / J. Zhong [et al.] / Comb Chem High Throughput Screen.
Tackling the poor assumptions of naive bayes text classifiers / J.D. Rennie [et al.] // Proceedings of the Twentieth International Conference on Machine Learning (ICML-2003). ‒ Washington DC, 2003. ‒ P. 616–623.
Hastie, T. The Elements of Statistical Learning / T. Hastie, R. Tibshirani, J. Friedman. ‒ Springer, 2009. ‒ P. 106–119.
Guyon, I. Automatic Capacity Tuning of Very Large VC-dimension Classifiers / I. Guyon, B. Boser, V. Vapnik // Advances in neural information processing. ‒ 1993.
Cortes, C. Support-vector networks / C. Cortes, V. Vapnik // Machine Learning. ‒ 1995. – V. 20. ‒ P. 273–297.
Classification and Regression Trees / L. Breiman [et al.] // Wadsworth, Belmont-CA. ‒ 1984.
Scikit-learn: Machine Learning in Python / Pedregosa [et al.] // JMLR. ‒ 2011. – V. 12, ‒ P. 2825–2830.
Machine learning models in breast cancer survival prediction / M. Montazeria [et al.] // Technology and Health Care. –2016.
Artificial Neural Networks in Image Processing for Early Detection of Breast Cancer / M.M. Mehdy [et al.] // Comput Math Methods Med. – 2017.
Assessing Breast Cancer Risk with an Artificial Neural Network / M. Sepandi [et al.] // Comput Math Methods Med. –2017.
Application of artificial neural network model combined with four biomarkers in auxiliary diagnosis of lung cancer / Xiaoran Duan [et al.] // Med Biol Eng Comput. – 2017. – № 55, V. 8. – P. 1239–1248.
Korhani Kangi, A. Predicting the Survival of Gastric Cancer Patients Using Artificial and Bayesian Neural Networks / A. Korhani Kangi, A. Bahrampour // Asian Pac J Cancer Prev. – 2018. – № 19, V. 2. – P. 487–490.
Delen, D. Predicting breast cancer survivability: a comparison of three data mining methods / D. Delen, G. Walker, A. Kadam // Artificial intelligence in medicine. – 2005. – № 34, V. 2. – P. 113–127.
Jiang, W. A naive Bayes algorithm for tissue origin diagnosis (TOD-Bayes) of synchronous multifocal tumors in the hepatobiliary and pancreatic system / W. Jiang [et al.] // Int J Cancer. – 2018. – № 42, V. 2. – P. 357–368.
Maniruzzaman, Md. Comparative Approaches for Classification of Diabetes Mellitus Data: Machine Learning Paradigm / Md. Maniruzzaman // Computer Methods and Programs in Biomedicine. – 2017. – V. 152. – P. 23–34.
Applications of Support Vector Machine (SVM) Learning in Cancer Genomics / S. Huang [et al.] // Cancer Genomics Proteomics. – 2018. – № 15, V. 1. – P. 41–51.
Viswanath, S.E. Comparing radiomic classifiers and classifier ensembles for detection of peripheral zone prostate tumors on T2-weighted MRI: a multi-site study / S.E.Viswanath [et al.] // BMC Med Imaging. – 2019. – № 19, V. 1. – P. 22–34.
Predictors of the therapeutic effect of corticosteroids on radiation-induced optic neuropathy following naso-pharyngeal carcinoma/ B. Zheng [and et.] // Support Care Cancer. – 2019.
Prostate cancer prediction using the random forest algorithm that takes into account transrectal ultrasound findings, age, and serum levels of prostate-specific antigen / L.H. Xiao [et al.] // Asian J Androl. – 2017. – № 19, V. 5. – P. 586–590.
Camp, R.L. Validation of tissue microarray technology in breast carcinoma / R.L. Camp, L.A. Charette, D.L. Rimm // Lab Invest. – 2000. – № 80, V. 12. – P. 1943–1949.
Zhang, H. The optimality of Naive Bayes / H. Zhang // Proc. FLAIRS. ‒ 2004. – № 1, V. 2. ‒ P. 1–6.
Kohavi, R. A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection / R. Kohavi // Intl. Jnt. Conf. AI Montreal, Quebec, Canada. ‒ 1995. ‒ V. 2 ‒ P. 1137–1143.