Using Urine Metabolome Profiles to Diagnose Breast Cancer
Author Information
Author(s): Kim Younghoon, Koo Imhoi, Jung Byung Hwa, Chung Bong Chul, Lee Doheon
Primary Institution: KAIST
Hypothesis
Multivariate classification methods can identify urinary biomarkers for breast cancer that univariate methods cannot.
Conclusion
Multivariate classifications are essential for accurately diagnosing breast cancer and identifying potential urinary biomarkers.
Supporting Evidence
- Five potential urinary biomarkers for breast cancer were identified with high accuracy.
- Four of the biomarkers were not identifiable using univariate methods.
- Multivariate methods showed better performance than univariate methods by 6.6-12.7 percent.
Takeaway
Scientists found that looking at many urine markers together helps to spot breast cancer better than just looking at one marker at a time.
Methodology
The study used multivariate classification techniques to analyze urine metabolome data and identify potential biomarkers for breast cancer.
Potential Biases
Potential biases may arise from the sample selection and the need for further experimental confirmation.
Limitations
Further validation with independent cohorts is needed to confirm the findings.
Participant Demographics
50 female breast cancer patients and 50 healthy controls, matched for age.
Statistical Information
P-Value
2.866e-06
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website