Classification for Urinary Bladder Epithelial Cancer Cell Basing on Random Forest Algorithm
Objectives: Classification diagnosis of urinary bladder epithelial cancer cell was strongly influenced by subjective judgments, which is made by pathologists, resulting in the presence of high false positive rate and false negative rate of diagnosis. Therefore, Random Forest was performed to diagnosis and classification urinary bladder epithelial cancer cell for exploring the feasibility and application value of its method. Methods: A total number of 258 urinary bladder epithelial cancer samples were collected and diagnosed. Morphological and colorimetric features of samples were evaluated by the application of ImageJ. Random Forest algorithm, intergrated with Weka 3.6.6 was performed to training samples and modeling. Test accuracy was calculated by 10-fold Cross-validation. Results: The overall classification accuracy performed by random forest was 98.13% between normal group and lesions group, 98.95% between urothelium dysplastic exfoliated cells and bladder urothelial cancer exfoliated cells. For the classification diagnosis of urinary bladder epithelial cancer cell, the classification diagnostic effect performed by Random Forest was the best while distinguishing lesions cells from normal cells, and bladder urothelial cancer exfoliated cells from urothelium dysplastic exfoliated cells, respectively. Conclusion: It was indicated that Random Forest can be considered as an effective classification method to classified urinary bladder epithelial cancer cells.
1.Earl J, Rico D, Carrillodesantapau E, et al. The UBC-40 Urothelial Bladder Cancer cell line index: a genomic resource for functional studies. BMC Genomics 2015;16:403. [CrossRef ]
2. Liu J, Qian D, He L, et al. PinX1 suppresses bladder urothelial carcinoma cell proliferation via the inhibition of telom erase activity and p16/cyclin D1 pathway. Molecular Cancer 2013;12:148. [CrossRef ]
3. Paner G P, Barkan G A, Mehta V, et al. Urachal carcinomas of the nonglandular type: salient features and considerations in pathologic diagnosis. The American Journal of Surgical Pathology 2012;36:432–42. [CrossRef ]
4. Dancik G M, Ru Y, Owens C, et al. A Framework to Select Clinically Relevant Cancer Cell Lines for Investigation by Establishing Their Molecular Similarity with Primary Human Cancers. Cancer Research 2011;71:7398–409. [CrossRef ]
5. Lopezbeltran A, Amin M B, Oliveira P, et al. Urothelial carcinoma of the bladder, lipid cell variant: clinicopathologic findings and LOH analysis. The American Journal of Surgical Pathology 2010;34:371–6. [CrossRef ]
6. Pascal L E, Deutsch E, Campbell D, et al. The urologic epithelial stem cell database (UESC) - a web tool for cell type-specific gene expression and immunohistochemistry images of the prostate and bladder. BMC Urology 2007;7. [CrossRef ]
7. Tachibana M, Miyakawa A, Nakashima J, et al. Constitutive production of multiple cytokines and a human chorionic gonadotrophin beta-subunit by a human bladder cancer cell line (KU-19-19): possible demonstration of totipotential differentiation. British Journal of Cancer 1997;76:163–74.
8. Yao W J, Chang C J, Chan S H, et al. Significance of urinary tissue polypeptide specific antigen (TPS) determination in patients with urothelial carcinoma. Anticancer Research 1995:2819–23.
9. Burchill S A, Bradbury M, Pittman K, et al. Detection of epithelial cancer cells in peripheral blood by reverse transcriptase-polymerase chain reaction. British Journal of Cancer 1995;278–81. [CrossRef ]
10. Sinard J H, Macleay L, Melamed J, et al. Hepatoid adenocarcinoma in the urinary bladder: unusual localization of a newly recognized tumor type. Cancer 1994;73:1919–25. [CrossRef ]
11. Yang L, Zhu Y T, Li-Li M A, et al. Characteristics of bladder transitional cell carcinoma with E-cadherin and N-cadherin double-negative expression. Oncology Letters 2016;12:530– 6. [CrossRef ]
12. Sasaki H, Yoshiike M, Nozawa S, et al. Expression Level of Urinary MicroRNA-146a-5p Is Increased in Patients With Bladder Cancer and Decreased in Those After Transurethral Resection. Clinical Genitourinary Cancer 2016;14. [CrossRef ]
13. Berrondo C, Flax J, Kucherov V, et al. Expression of the Long Non-Coding RNA HOTAIR Correlates with Disease Progression in Bladder Cancer and Is Contained in Bladder Cancer Patient Urinary Exosomes. Plos One 2016;11. [CrossRef ]
14. Syed I S, Pedram A, Farhat W A. Role of Sonic Hedgehog (Shh) Signaling in Bladder Cancer Stemness and Tumorigenesis. Current Urology Reports 2016;17:1–7. [CrossRef ]
15. Chiu K Y, Wu C C, Chia C H, et al. Inhibition of growth, migration and invasion of human bladder cancer cells by an-trocin, a sesquiterpene lactone isolated from Antrodia cinnamomea, and its molecular mechanisms. Cancer Letters 2015;373:174–84. [CrossRef ]
16. Hu H, Zhao J, Zhang M. Expression of Annexin A2 and Its Correlation With Drug Resistance and Recurrence of Bladder Cancer. Technology in Cancer Research & Treatment 2015.
17. Tan J, Qiu K, Li M, et al. Double‐negative feedback loop between long non‐coding RNA TUG1 and miR‐145 promotes epithelial to mesenchymal transition and radioresistance in human bladder cancer cells. Febs Letters 2015;589:3175–81.
18. Geng H, Zhao L, Liang Z, et al. ERK5 positively regulates cigarette smoke-induced urocystic epithelial-mesenchymal transition in SV 40 immortalized human urothelial cells. Oncology Reports 2015;34:1581–8. [CrossRef ]