Understanding Prostate Cancer Risk Using Statistical and Machine Learning Approaches - A Comparative Methodological Analysis

Aug 2, 2026

Upload by Eden

Prostate cancer represents one of the most prevalent and lethal malignancies among men worldwide, requiring accurate risk prediction tools for earlier detection and personalized clinical care. This comparative study evaluates traditional binary logistic regression against machine learning models—support vector machines, K-nearest neighbors, CHAID, and C5.0—using demographic, clinical, and lifestyle data from 501 participants. Results show age, smoking, and family history as consistent predictors, with logistic regression achieving the highest accuracy (92.2%).

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Abstract

Materials and Methods

Results

Conclusion

Introduction

Hamidiye Med J 2025;6(3):171-177 Understanding Prostate Cancer Risk Using Statistical and Machine Learning Approaches: A Comparative Methodological Analysis İstatistiksel ve Makine Öğrenmesi Yaklaşımlarını Kullanarak Prostat Kanseri Riskini Anlamak: Karşılaştırmalı MetodolojikAnaliz Selman Aktaş1, Murat Kirişci2, Muzaffer Akçay3, Muhammet Çiçek4 1University of Health Sciences Türkiye, Hamidiye Faculty of Medicine, Department of Biostatistics and Medical Informatics, İstanbul, Türkiye 2İstanbul University-Cerrahpaşa, Cerrahpaşa Faculty of Medicine, İstanbul, Türkiye 3Bezmialem Vakıf University Faculty of Medicine, Department of Urology, İstanbul, Türkiye 4İstanbul Medeniyet University Faculty of Medicine, Department of Urology, İstanbul, Türkiye ABSTRACT Background: Prostate cancer is one of the most common and lethal malignancies among men worldwide, making accurate risk prediction tools essential for early diagnosis and personalized care. This study aimed to compare the predictive ability of traditional binary logistic regression with machine learning (ML) algorithms, including support vector machines (SVM), K-nearest neighbors (KNN), chi-squared automatic interaction detection (CHAID), and C5.0, in identifying key risk factors and classifying prostate cancer status. Materials and Methods: The study included 501 male participants (248 diagnosed cases and 253 controls) who completed a structured 20-item questionnaire covering demographic, clinical, and lifestyle characteristics. Results: Age, smoking status, and family history of cancer consistently emerged as significant predictors across models. Additional indicators included blood in semen or urine, frequency of urination, and daily activity level. Logistic regression achieved the highest accuracy (92.2%), followed by CHAID (91.36%), SVM (89.92%), KNN (88.48%), and C5.0 (88%). Conclusion: Logistic regression provided the best accuracy and interpretability for structured clinical data, while ML models offered complementary insights by identifying complex, nonlinear associations. Keywords: Prostate cancer, risk prediction, logistic regression, machine learning, classification algorithms ÖZ Amaç: Prostat kanseri, erkekler arasında en yaygın ve ölümcül malignitelerden biridir. Erken tanı ve kişiselleştirilmiş bakım içindoğru risk tahmin araçlarının geliştirilmesi büyük önem taşır. Bu çalışmada, prostat kanseri risk faktörlerini belirleme ve hastalıkdurumunu sınıflandırmada ikili lojistik regresyon ile makine öğrenimi (ML) algoritmalarının (SVM, KNN, CHAID ve C5.0) öngörü performansları karşılaştırılmıştır. Gereç ve Yöntemler: Çalışmaya, demografik, klinik ve yaşam tarzı özelliklerini içeren 20 soruluk yapılandırılmış anketi dolduran 501 erkek (248 hasta ve 253 kontrol) dahil edilmiştir. Bulgular: Yaş, sigara kullanımı ve ailede kanser öyküsü tüm modellerde anlamlı öngörücüler olarak bulunmuştur. Ek olarak semen veya idrarda kan, idrara çıkma sıklığı ve günlük aktivite düzeyi de belirleyici olmuştur. Lojistik regresyon %92,2 doğrulukla en yüksek performansı göstermiştir. CHAID %91,36, SVM %89,92, KNN %88,48 ve C5.0 %88 doğruluk oranına ulaşmıştır. Sonuç: Lojistik regresyon yapılandırılmış klinik verilerde en yüksek doğruluk ve yorumlanabilirliği sağlarken, ML algoritmaları karmaşık ve doğrusal olmayan ilişkileri ortaya çıkararak tamamlayıcı katkılar sunmuştur. Anahtar Kelimeler: Prostat kanseri, risk tahmini, lojistik regresyon, makine öğrenmesi, sınıflandırma algoritmaları Address for Correspondence: Selman Aktaş, University of Health Sciences Türkiye, Hamidiye Faculty of Medicine, Department of Biostatistics and Medical Informatics,İstanbul, Türkiye E-mail: [selmanakts@gmail.com](selmanakts@gmail.com ORCID ID: orcid.org/0000-0002-8493-5000)[ ORCID ID:](selmanakts@gmail.com ORCID ID: orcid.org/0000-0002-8493-5000)[ orcid.org/0000-0002-8493-5000](selmanakts@gmail.com ORCID ID: orcid.org/0000-0002-8493-5000) Received: 29.04.2025 Accepte

FAQs

Which models were compared for prostate cancer risk prediction in this study?

The study compared binary logistic regression with machine learning algorithms including SVM, KNN, CHAID, and C5.0 to classify prostate cancer status.

What participant information was used to build the predictive models?

Data came from a structured 20-item questionnaire covering demographic, clinical, and lifestyle characteristics.

What factors were consistently significant across the models?

Age, smoking status, and family history of cancer emerged as significant predictors across the evaluated models.

How did logistic regression perform compared with the machine learning methods?

Logistic regression delivered the highest accuracy at 92.2%, outperforming CHAID (91.36%), SVM (89.92%), KNN (88.48%), and C5.0 (88%).

Aug 2, 2026

Upload by Eden