Using Neural Networks to Detect Ovarian Cancer Early
Author Information
Author(s): Thakur Ankita, Mishra Vijay, Jain Sunil K.
Primary Institution: Indian Institute of Technology Bombay
Hypothesis
Can SELDI-TOF MS profiling of plasma proteins coupled with artificial intelligence distinguish between normal controls and patients with malignant ovarian cancer?
Conclusion
The study found that data mining techniques can effectively distinguish ovarian cancer patients from healthy individuals using serum proteomic patterns.
Supporting Evidence
- The method achieved 100% training accuracy and high testing accuracy of 99.16% and 98.50%.
- The study demonstrated that the neural network could effectively classify ovarian cancer patients from healthy controls.
- Significant features for classification were identified using student t-test.
Takeaway
Scientists used a special computer program to look at blood samples and found a way to tell if someone has ovarian cancer early.
Methodology
The study used SELDI-TOF MS for serum proteomic analysis and applied student t-test and neural networks for classification.
Potential Biases
Potential bias due to the limited sample size and selection of datasets.
Limitations
The study only used two datasets, which may limit the generalizability of the findings.
Participant Demographics
The sample included 121 ovarian cancer patients and 95 healthy controls.
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