Related Experiment Video
Updated: May 14, 2025
13:24
Integration of Wet and Dry Bench Processes Optimizes Targeted Next-generation Sequencing of Low-quality and Low-quantity Tumor Biopsies
Published on: April 11, 2016
11.7K
Towards precision oncology: a multi-level cancer classification system integrating liquid biopsy and machine learning
Amr Eledkawy1, Taher Hamza1, Sara El-Metwally2,3
1Department of Computer Science, Faculty of Computers and Information, Mansoura University, P.O. Box: 35516, Mansoura, Egypt.
Biodata Mining
|April 11, 2025
Summary
This study introduces a novel multi-level cancer classification system using plasma biomarkers for early detection of seven cancer types. The system achieves high accuracy, improving patient outcomes through timely interventions.
Area of Science:
- Biomarker Discovery
- Computational Oncology
- Precision Medicine
Background:
- Cancer remains a leading cause of mortality worldwide.
- Early detection significantly improves patient survival rates and treatment efficacy.
- This study focuses on identifying seven common cancer types.
Purpose of the Study:
- To develop a multi-level cancer classification system.
- To utilize plasma cell-free DNA (cfDNA)/circulating tumor DNA (ctDNA) mutations and protein biomarkers.
- To enhance early cancer detection and improve patient outcomes.
Main Methods:
- A multi-stage binary classification framework tailored for specific cancer types.
- Majority vote feature selection combining six methods (Information Value, Chi-Square, RF, ET, RFE, L1).
- Customized classifiers (XGBoost, RF, ET, QDA) and ensemble soft voting for optimized accuracy.
Main Results:
- The system achieved an Area Under the Curve (AUC) of 98.2% and an accuracy of 96.21%.
- Performance surpassed previously published results.
- Models and dataset are publicly available on GitHub for reproducibility.
Conclusions:
- Identified biomarkers improve diagnostic interpretability and aid decision-making.
- The system demonstrates high effectiveness in tissue localization.
- Contributes to better patient outcomes via timely and precise medical interventions.
