Accès ouvert · CC BY

A Hybrid Model Integrating Interval Type-2 Fuzzy Logic (IT2FL) and XGBoost for Enhanced Prostate Cancer Analytics

Article scientifique 2026 Sciences de la santé Anglais

Résumé

Diagnosing prostate cancer continues to be difficult because of the drawbacks related to traditional methods like PSA testing and digital rectal examination, which can yield false positives, false negatives, and subjective results. These issues may lead to avoidable biopsies and postponed recognition of high-risk situations. This research introduces a hybrid diagnostic framework that combines eXtreme Gradient Boosting (XGBoost) with Interval Type-2 Fuzzy Logic (IT2FL) to tackle uncertainty in assessing prostate cancer risk. The IT2FL component represents uncertainty in clinical information with linguistic rules and assesses three primary factors: Age, Family History, and Race (African Ancestry) using 18 fuzzy rules defined by experts to produce a binary risk classification (Yes/No). The resulting representations are then handled by XGBoost for the final classification. The evaluation of performance on 5,589 test samples resulted in a total accuracy of 0.97. For the No class, the model reached a precision of 0.97, a recall of 0.83, and an F1-score of 0.89, whereas the Yes class attained a precision of 0.96, a recall of 0.99, and an F1-score of 0.98. The macro-average precision, recall, and F1-score were 0.97, 0.91, and 0.94, respectively, while the weighted-average F1-score stood at 0.96. An evaluation of the case with Age = 52, Family History = 0, and Race = 1 further illustrated the model's use in personal risk categorization. The results show that the IT2FL-XGBoost framework delivers classified capabilities while maintaining the clarity of fuzzy reasoning, highlighting its potential as a decision-support method for assessing prostate cancer risk.

Citer ce document

Umoh, U. A., Inyang, U. G., E., A. B., & Ubong, E. A. (2026). A Hybrid Model Integrating Interval Type-2 Fuzzy Logic (IT2FL) and XGBoost for Enhanced Prostate Cancer Analytics. https://doi.org/10.68050/jams.2026.741

Exporter : BibTeX · RIS (Zotero, Mendeley, EndNote)

Accès au document

Texte intégral archivé sur SPHAERO, en accès ouvert : lisible gratuitement en ligne après connexion. Le document reste aussi disponible à sa source. Se connecter

Voir l'article sur le site de la revue

Licence et provenance

Licence : CC BY

Notice moissonnée depuis OpenAlex le 16/11/2026. Le document reste hébergé par sa source.
Voir le document à la source →

Statistiques

Consultations (hors robots) : 2

Téléchargements (hors robots) : 0

Méthode de comptage