EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Machine Learning-Based Predictive Model for Early Disease Detection

Sakshi, Divya Singla, Nisha, Anisha

International Journal of Research and Review in Applied Science, Humanities, and Technology · 28.12.2024 · 227-233

DOI: 10.71143/ghagbv94 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Research and Review in Applied Science, Humanities, and Technology, online erschienen am 28.12.2024. Verfasst von Sakshi und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Detecting a disease at an early stage improves the chance of effective treatment, but the decision depends on many correlated clinical measurements. This paper develops and evaluates a machine-learning predictive model for early disease detection and reports the full evaluation pipeline, including the safeguards against optimistic results. Eight classifiers (logistic regression, k-NN, SVM, naive Bayes, decision tree, random forest, gradient boosting and a neural network) are tuned by randomised search and compared by nested cross-validation on the public Wisconsin Diagnostic Breast Cancer dataset (569 patients, 30 features). The RBF-kernel SVM ranks first with ROC-AUC 99.6%, accuracy 97.8% and sensitivity 96.0%; the four best models are statistically close (Wilcoxon p = 0.051 between the top two). On a held-out test set of 114 patients the SVM reaches accuracy 98.2% and AUC 99.6%, with wide confidence intervals because of the small sample. A threshold tuned for 98% training sensitivity did not improve test sensitivity and cost specificity, and a feature-count ablation shows that 10 features lose about 0.7 AUC points against all 30. The results are from one public dataset without external validation and are not a clinical claim.

Zitieren

APA

Sakshi, Singla, D., Nisha, & Anisha (2024). Machine Learning-Based Predictive Model for Early Disease Detection. International Journal of Research and Review in Applied Science, Humanities, and Technology, 227-233. https://doi.org/10.71143/ghagbv94

Vancouver

Sakshi, Singla D, Nisha, Anisha. Machine Learning-Based Predictive Model for Early Disease Detection. International Journal of Research and Review in Applied Science, Humanities, and Technology. 2024:227-233. doi:10.71143/ghagbv94

BibTeX
@article{Sakshi2024,
  author = {Sakshi and Singla, Divya and Nisha and Anisha},
  title = {Machine Learning-Based Predictive Model for Early Disease Detection},
  journal = {International Journal of Research and Review in Applied Science, Humanities, and Technology},
  year = {2024},
  pages = {227--233},
  doi = {10.71143/ghagbv94}
}
Kostenlos starten →