Optimized Support Vector Machine for Lung Cancer Classification
Author Information
Author(s): Oyediran Mayowa O., Ojo Olufemi S., Raji Ibrahim A., Adeniyi Abidemi Emmanuel, Aroba Oluwasegun Julius
Primary Institution: Ajayi Crowther University, Oyo, Nigeria
Hypothesis
The study aims to enhance machine learning to increase the precision and quality of lung cancer classification.
Conclusion
The developed chameleon swarm-based SVM technique significantly improved lung cancer classification performance compared to traditional methods.
Supporting Evidence
- The CS-SVM approach achieved 97.33% accuracy, outperforming the traditional SVM which had 95.90% accuracy.
- CS-SVM reduced the false-positive rate to 3.57% compared to 5.95% for SVM.
- CS-SVM demonstrated higher sensitivity (97.71%) than SVM (96.69%) in identifying malignant nodules.
- CS-SVM also improved specificity to 96.43% from SVM's 94.05%.
Takeaway
This study created a smart computer program that helps doctors tell if lung nodules are cancerous or not, using special math and pictures from CT scans.
Methodology
The study used a dataset of CT scan images, applying image processing techniques and a chameleon swarm optimization algorithm to enhance SVM performance.
Limitations
The study uses a restricted dataset that may not fully represent the variability in clinical settings.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website