Related Experiment Video
Updated: Jan 18, 2026
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
7.4K
Effectiveness of Multi-Layer Perceptron-Based Binary Classification Neural Network in Detecting Breast Cancer Through
Eun-Gyeong Lee1, Jaihong Han1, Seeyoun Lee1
1Department of Surgery, Center of Breast Cancer, Research Institute and Hospital, National Cancer Center, Goyang 10408, Republic of Korea.
Cancers
|September 13, 2025
Summary
A new nine-protein blood test significantly improves breast cancer detection accuracy compared to an older three-protein test. This enhanced diagnostic tool shows higher sensitivity and specificity for early cancer identification.
Area of Science:
- Biomarker Discovery
- Medical Diagnostics
- Artificial Intelligence in Medicine
Background:
- An existing three-protein signature is used for blood-based breast cancer diagnostics.
- A novel nine-protein serum signature has been developed to improve diagnostic accuracy.
Purpose of the Study:
- To evaluate the clinical sensitivity and specificity of a medical device using the nine-protein signature.
- To assess the performance of an artificial intelligence algorithm for breast cancer diagnosis.
Main Methods:
- Quantified nine specific proteins (APOC1, CHL1, FN1, VWF, PPBP, CLU, PRDX6, PRG4, MMP9) in serum samples.
- Utilized multiple reaction monitoring via mass spectrometry for protein quantification.
- Analyzed samples from 243 healthy controls and 222 breast cancer patients.
Main Results:
- The nine-protein signature achieved 83.3% sensitivity and 88.1% specificity for breast cancer diagnosis.
- The three-protein signature showed 71.6% sensitivity and 85.3% specificity.
- The nine-protein signature demonstrated higher accuracy (85.8%) compared to the three-protein signature (77.0%).
Conclusions:
- The nine-protein signature offers a more accurate and effective method for breast cancer detection.
- This enhanced signature shows potential for improving blood-based cancer diagnostics.
- Artificial intelligence models can effectively utilize protein signatures for clinical assessment.
