PriMera Scientific Medicine and Public Health (ISSN: 2833-5627)

Research Article

Volume 5 Issue 1

Modeling the Factors Affecting Gleason Score with Artificial Neural Networks and Indirectly Determining Prostate Cancer Risk Factors

Zeynep Kucukakcali* and Ipek Balikci Cicek

June 27, 2024

View PDF

Abstract

Aim: Prostate cancer, one of the most common cancer among men and a cancer that can vary significantly in its aggressiveness, will cause more deaths in the future with its increasing incidence. Gleason score has been defined as the most reliable and autonomous predictor of prostate cancer outcomes. The study aim to determine the variables affecting Gleason score and indirectly to establish prognostic indicators for prostate cancer.

Material and methods: The variables in the data set were analyzed according to the dependent variable categories Independent sample test and Mann Whitney U test were used in statistical analyses and p<0.05 was considered significant. Analyses were performed with IBM SPSS 26.0. In the modeling phase, the relationship between the grouped form of Gleason score and other variables was examined with Multilayer Perceptron and Radial Basis Function Neural Network methods. The dataset was divided into training and test datasets in a 70:30 ratio. The results are reported using accuracy, balanced accuracy, sensitivity, specificity, PPV, NPV and F1 score as performance metrics.

Results: The data set used in the study consists of variables belonging to 97 patients. The mean age of the patients was 63.87 years. Patients were divided into two groups: those with a Gleason value of 7 and above and those with a Gleason value below 7. There were 35 patients with a Gleason value below 7 and 62 patients with a Gleason value above 7. According to the results obtained from the modeling, the best result was obtained from the Multilayer Perceptron model. accuracy, balanced accuracy, sensitivity, specificity, positive predictive value, negative predictive value and F1 score were 96%, 100%, 93.3%, 90.9%, 100%, 95.2%, respectively.

Conclusion: The study obtained highly accurate classification results when modeling with Gelason score categories and other independent variables. This shows that machine learning models can be used effectively and successfully in medical data. Furthermore, important variables were identified and their indirect associations with prostate cancer were revealed. In the future, more detailed research on prostate cancer can be conducted by focusing on these variables.

Keywords: Prostate cancer; Gleason score; modelling; classification

References

  1. Habib A., et al. “Risk factors associated with prostate cancer”. Journal of Drug Delivery and Therapeutics 11.2 (2021): 188-93.
  2. Culp MB., et al. “Recent Global Patterns in Prostate Cancer Incidence and Mortality Rates”. European urology 77.1 (2020): 38-52.
  3. Rawla P. “Epidemiology of prostate cancer”. World journal of oncology 10.2 (2019): 63-89.
  4. Archer M, Dogra N and Kyprianou N. “Inflammation as a driver of prostate cancer metastasis and therapeutic resistance”. Cancers 12.10 (2020): 2984.
  5. Tınay İ and Türkeri L. “A New Era in Metastatic Prostate Cancer: “The Combination of Chemotherapy and Hormonal Treatment as Initial Treatment”. Bulletin of Urooncology (2016).
  6. Bastian PJ., et al. “Characteristics of Insignificant Clinical T1c Prostate Tumors”. Cancer (2004): 2001-5.
  7. Mahal BA., et al. “Prostate Cancer-Specific Mortality Across Gleason Scores in Black vs Nonblack Men”. Jama (2018): 2479-2481.
  8. Stamey TA., et al. “Prostate Specific Antigen in the Diagnosis and Treatment of Adenocarcinoma of the Prostate. II. Radical Prostatectomy Treated Patients”. The Journal of Urology 141.5 (1989): 1076-83.
  9. Ramchoun H., et al. “Multilayer Perceptron: Architecture Optimization and Training”. International Journal of Interactive Multimedia and Artificial Intelligence (2016).
  10. Wang Z, Liu Z and Zheng CD. Introduction to Neural Networks (2015).
  11. Yi W, Zhao Y and Chan APC. “A Tailored Artificial Intelligence Model for Predicting Heat Strain of Construction Workers”. Iop Conference Series Earth and Environmental Science (2022).
  12. Tekin TG and Patir S. “Estimating the American Dollar Exchange Rate by Artificial Neural Networks: 2009 - 2021 Period”. R&s - Research Studies Anatolia Journal (2023).
  13. Yıldırım DK, Cemek B and Kucuktopcu E. “Daily Evaporation Estimation with Fuzzy Artificial Neural Networks and Multilayer Artificial Neural Networks”. Earth Water Magazine (2019).
  14. Çİçek İB and Küçükakçali Z. “Classification of prostate cancer and determination of related factors with different artificial neural network”. Middle Black Sea Journal of Health Science 6.3 (2020): 325-32.
  15. Zhou CK., et al. “Prostate Cancer Incidence in 43 Populations Worldwide: An Analysis of Time Trends Overall and by Age Group”. International Journal of Cancer 138.6 (2015): 1388-400.
  16. Djiwa T., et al. “Prostate Cancers in Men Under the Age of 50: About a Series in Togo, Sub-Saharan Africa”. BMC Cancer 22.1 (2022): 1341.
  17. Allott EH, Masko EM and Freedland SJ. “Obesity and Prostate Cancer: Weighing the Evidence”. European Urology 63.5 (2013): 800-9.
  18. Sheng T., et al. “No Association Between Fiber Intake and Prostate Cancer Risk: A Meta-Analysis of Epidemiological Studies”. World Journal of Surgical Oncology 13.1 (2015): 264.
  19. Allott EH and Hursting SD. “Obesity and Cancer: Mechanistic Insights from Transdisciplinary Studies”. Endocrine Related Cancer 22.6 (2015): R365-R86.
  20. Moradi A, Zamani M and Moudi E. “A Systematic Review and Meta-Analysis on Incidence of Prostate Cancer in Iran”. Health Promotion Perspectives 9.2 (2019): 92-8.
  21. McDonald MJ and Parsons JK. “The Case for Tailored Prostate Cancer Screening: An NCCN Perspective”. Journal of the National Comprehensive Cancer Network 13.12 (2015): 1576-83.
  22. Butler EN., et al. “Fatal Prostate Cancer Incidence Trends in the United States and England by Race, Stage, and Treatment”. British Journal of Cancer 123.3 (2020): 487-94.
  23. Porcacchia AS., et al. “Sleep Disorders and Prostate Cancer Prognosis: Biology, Epidemiology, and Association with Cancer Development Risk”. European Journal of Cancer Prevention 31.2 (2021): 178-89.
  24. Kikugawa T. “Primary Gleason Pattern Does Not Affect Recurrence-Free Survival in Patients Receiving Radiotherapy for Prostate Cancer”. Journal of Nuclear Medicine & Radiation Therapy 05.03 (2014).
  25. Sinha AA., et al. “Ratio of Cathepsin B to Stefin a Identifies Heterogeneity Within Gleason Histologic Scores for Human Prostate Cancer”. The Prostate 48.4 (2001): 274-84.
  26. Epstein JI., et al. “A Contemporary Prostate Cancer Grading System: A Validated Alternative to the Gleason Score”. European Urology 69.3 (2016): 428-35.
  27. Kweldam CF., et al. “Cribriform Growth Is Highly Predictive for Postoperative Metastasis and Disease-Specific Death in Gleason Score 7 Prostate Cancer”. Modern Pathology 28.3 (2015): 457-64.
  28. Ren J., et al. “Differentiation Among Prostate Cancer Patients with Gleason Score of 7 Using Histopathology Image and Genomic Data”. Proc SPIE Int Soc Opt Eng (2018).
  29. Epstein JI., et al. “The 2005 International Society of Urological Pathology (ISUP) consensus conference on Gleason grading of prostatic carcinoma”. The American journal of surgical pathology 29.9 (2005): 1228-1242.
  30. Humphrey PA. “Gleason grading and prognostic factors in carcinoma of the prostate”. Modern pathology 17.3 (2004): 292-306.
  31. Humphrey PA. “Prostate Pathology”. Chicago: ASCP Press (2003).
  32. Thompso  N Ian M., et al. “Prevalence of prostate cancer among men with a prostate-specific antigen level≤ 4.0 ng per milliliter”. New England Journal of Medicine 350.22 (2004): 2239-2246.