Breast Cancer Event Prediction Using Subtype-Specific Models
Author Information
Author(s): Sontrop Herman M. J., Verhaegh Wim F. J., Reinders Marcel J. T., Moerland Perry D.
Primary Institution: Philips Research, Eindhoven, The Netherlands
Hypothesis
Can subtype-specific predictors outperform untyped predictors in breast cancer event prediction?
Conclusion
Subtype-specific predictors significantly improve breast cancer event prediction compared to untyped predictors.
Supporting Evidence
- The study analyzed over 1500 arrays to develop subtype-specific predictors.
- Typed predictors showed higher performance metrics compared to untyped predictors.
- The methodology is applicable to other diseases beyond breast cancer.
Takeaway
This study shows that using specific types of breast cancer can help doctors make better predictions about how the disease will progress.
Methodology
The study used a novel experimental protocol to compare subtype-specific predictors with untyped predictors, controlling for sample size and class distributions.
Potential Biases
Potential bias due to unequal class distributions among subtypes.
Limitations
The study may not generalize to all breast cancer types due to the specific subtypes analyzed.
Participant Demographics
The study involved 892 breast cancer samples with varying subtypes.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website