Detecting Breast Cancer Subtypes with Raman Spectroscopy and Machine Learning
Author Information
Author(s): Tipatet Kevin Saruni, Hanna Katie, Davison‐Gates Liam, Kerst Mario, Downes Andrew
Primary Institution: Institute for BioEngineering, School of Engineering University of Edinburgh
Hypothesis
Can Raman spectroscopy combined with machine learning accurately classify breast cancer subtypes using blood plasma samples?
Conclusion
The study shows that combining Raman spectroscopy with machine learning can accurately classify the four major breast cancer subtypes at stage Ia with high sensitivity and specificity.
Supporting Evidence
- The method achieved an average sensitivity of 90% and specificity of 95%.
- The study is the first to classify breast cancer subtypes at stage Ia using Raman spectroscopy and machine learning.
- High AUC of 0.98 indicates excellent model performance.
- Statistical significance was confirmed using the Mann–Whitney–Wilcoxon test.
Takeaway
This study found a way to use a special light technique to tell different types of breast cancer apart using just a small sample of blood.
Methodology
The study used Raman spectroscopy to analyze blood plasma samples from breast cancer patients and healthy controls, applying machine learning for classification.
Potential Biases
Potential bias from sample selection and the small size of the pilot study.
Limitations
The study was limited to a small sample size and focused only on stage Ia breast cancer.
Participant Demographics
12 samples from stage Ia breast cancer patients and 12 from healthy volunteers.
Statistical Information
P-Value
p<0.05
Confidence Interval
99%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website