Journal of Artificial Intelligence and Digital Health

ABSTRACT

This study systematically reviews the integration of Interval Type-2 Fuzzy Logic (IT2FL) and XGBoost for prostate cancer detection and prognosis. Prostate cancer is a significant health concern, with existing diagnostic methods often challenged by data uncertainties and accuracy limitations. IT2FL effectively addresses uncertainties in medical data, while XGBoost enhances classification performance. Following PRISMA guidelines, a systematic search was conducted across eight sources including Google Scholar, IEEE Xplore, ACM Digital Library, Elsevier, Springer, ScienceDirect, ResearchGate, and Semantic Scholar yielding 500 records. After removing duplicates and applying exclusion criteria based on sample size, publication quality, geographic diversity, timeframe (2017–2024), language, and relevance to IT2FL–XGBoost approaches, 121 studies were retained for analysis. The review addresses four research aspects: the integration of IT2FL and XGBoost, input parameter significance, the role of linguistic variables, and the impact of XGBoost on diagnostic accuracy. Visualization tools such as VOSviewer and Matplotlib facilitated data analysis and presentation. Findings from the reviewed literature indicate that hybrid IT2FL–XGBoost approaches can enhance diagnostic precision, reduce false positives, and support clinical decision-making, though challenges such as data diversity and model interpretability persist across studies. This review synthesizes existing evidence on the feasibility of combining IT2FL and XGBoost for prostate cancer detection and prognosis, and identifies explainable AI integration and broader clinical validation as priorities for future research.

Keywords: Interval Type-2 Fuzzy Logic (IT2FL); XGBoost; Prostate Cancer; Prostate Cancer Detection; Systematic Review; Medical Decision Support Systems