A Novel Embedded Feature Selection and Dimensionality Reduction Method for an SVM Type Classifier to Predict Periventricular Leukomalacia (PVL) in Neonates
Source abstract
Abstract
This paper is concerned with the prediction of the occurrence of periventricular leukomalacia (PVL) in neonates after heart surgery. Our prior work shows that the Support Vector Machine (SVM) classifier can be a powerful tool in predicting clinical outcomes of such complicated and uncommon diseases, even when the number of data samples is low. In the presented work, we first illustrate and discuss the shortcomings of the traditional automatic machine learning (aML) approach. Consequently, we describe our methodology for addressing these shortcomings, while utilizing the designed interactive ML (iML) algorithm. Finally, we conclude with a discussion of the developed method and the results obtained. In sum, by adding an additional (Genetic Algorithm) optimization step in the SVM learning framework, we were able to (a) reduce the dimensionality of an SVM model from 248 to 53 features, (b) increase generalization that was confirmed by a 100% accuracy assessed on an unseen testing set, and (c) improve the overall SVM model’s performance from 65% to 100% testing accuracy, utilizing the proposed iML method.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Congenital Heart Disease Studies · Neonatal and fetal brain pathology · Phonocardiography and Auscultation Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Villanova Center for Analytics of Dynamic Systems, Villanova University, 800 Lancaster Ave, Villanova, PA 19085, USA
June and Steve Wolfson Laboratory for Clinical and Biomedical Optics, Children’s Hospital of Philadelphia, 324 S 34th St, Philadelphia, PA 19104, USA
Funding
National Institutes of Health · Villanova University
References (32)
Show source-provided references
1. Holzinger, A. (2014). Biomedical Informatics: Descovering Knowledge in Big Data, Springer International Publishing Switzerland. [1st ed.].10.1007/978-3-319-04528-3 ↗
2. Holzinger, A. (2016). Machine Learning for Health Informatics, Springer.10.1007/978-3-319-50478-0 ↗
3. A machine learning algorithm to improve patient-centric pediatric cardiopulmonary resuscitation10.1016/j.imu.2020.100339 ↗
4. Evolutionary computing for knowledge discovery in medical diagnosis10.1016/S0933-3657(03)00002-2 ↗
5. Interactive machine learning for health informatics: When do we need the human-in-the-loop?10.1007/s40708-016-0042-6 ↗
6. Neurocognitive monitoring and care during pediatric cardiopulmonary bypass-current and future directions10.2174/157340308784245766 ↗
7. Vapnik, V. (1998). Statistical Learning Theory, Springer.
8. Brain maturation is delayed in infants with complex congenital heart defects10.1016/j.jtcvs.2008.10.025 ↗
9. Prediction of periventricular leukomalacia. Part II: Selection of hemodynamic features using computational intelligence10.1016/j.artmed.2008.12.004 ↗
10. Preoperative cerebral blood flow is diminished in neonates with severe congenital heart defects10.1016/j.jtcvs.2004.07.022 ↗
11. Prediction of periventricular leukomalacia. Part I: Selection of hemodynamic features using logistic regression and decision tree algorithms10.1016/j.artmed.2008.12.005 ↗
12. Scoring system for periventricular leukomalacia in infants with congenital heart disease10.1038/pr.2015.99 ↗
13. Jalali, A., Berg, R., Nadkarni, V., and Nataraj, C. (2013, January 21–23). Improving Cardiopulmonary Resuscitation (CPR) by Dynamic Variation of CPR Parameters. Proceedings of the Dynamic Systems and Control Conference, ASME, Palo Alto, CA, USA.10.1115/DSCC2013-3879 ↗
14. Prediction of Periventricular Leukomalacia in Neonates after Cardiac Surgery Using Machine Learning Algorithms10.1007/s10916-018-1029-z ↗
15. Bender, D., Jalali, A., Licht, D.J., and Nataraj, C. (2015, January 28–30). Prediction of periventricular leukomalacia occurrence in neonates using a novel support vector machine classifier optimization method. Proceedings of the ASME 2015 Dynamic Systems and Control Conference, Columbus, OH, USA.
16. Bender, D., Jalali, A., Licht, D.J., and Nataraj, C. (2014, January 17–19). Prediction of periventricular leukomalacia occurrence in neonates using a novel unsupervised learning method. Proceedings of the Dynamics Systems and Control Conference, Washington, DC, USA.10.1115/DSCC2014-6304 ↗
17. Prediction of periventricular leukomalacia occurrence in neonates after heart surgery10.1109/JBHI.2013.2285011 ↗
18. Jalali, A., Licht, D.J., and Nataraj, C. (September, January 28). Application of decision tree in the prediction of periventricular leukomalacia (PVL) occurrence in neonates after heart surgery. Proceedings of the 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, CA, USA.10.1109/EMBC.2012.6347344 ↗
19. Settles, B. (2012). Active Learning. Synthesis Lectures on Artificial Intelligence and Machine Learning, Morgan and Claypool Publishers.10.1007/978-3-031-01560-1 ↗
20. Holzinger, A. (2016). Survey on Feature Extraction and Applications of Biosignals. Machine Learning for Health Informatics: State-of-the-Art and Future Challenges, Springer.10.1007/978-3-319-50478-0 ↗
21. Neurodevelopmental Outcomes in Children with Congenital Heart Disease—What Can We Impact?10.1097/PCC.0000000000000800 ↗
22. Flach, P. (2012). Machine Learning—The Art and Science of Algorithms that Make Sense of Data, Cambridge University Press. [1st ed.].10.1017/CBO9780511973000 ↗
23. Russell, S.J., and Norvig, P. (2010). Artificial Intelligence a Modern Approach, Pearson Higher Education. [3rd ed.].
24. Saghapour, E., Kermani, S., and Sehhati, M. (2017). A novel feature ranking method for prediction of cancer stages using proteomics data. PLoS ONE, 12.10.1371/journal.pone.0184203 ↗
25. Wrappers for feature subset selection10.1016/S0004-3702(97)00043-X ↗
26. Wang, G., Lochovsky, F.H., and Yang, Q. (2004, January 1–5). Feature Selection with Conditional Mutual Information MaxiMin in Text Categorization. Proceedings of the Thirteenth ACM International Conference on Information and Knowledge Management 2004, Queensland, Australia.10.1145/1031171.1031241 ↗
27. Novovicova, J., Somol, P., Haindl, M., and Pudil, P. (2007, January 7–10). Conditional Mutual Information Based Feature Selection for Classification Task. Proceedings of the Iberoamerican Congress on Pattern Recognition 2007, Valparaíso, Chile.10.1007/978-3-540-76725-1_44 ↗
28. A review of feature selection methods based on mutual information10.1007/s00521-013-1368-0 ↗
29. A novel community detection based genetic algorithm for feature selection10.1186/s40537-020-00398-3 ↗
30. A Tutorial on Support Vector Machines for Pattern Recognition10.1023/A:1009715923555 ↗
31. Xie, Y., and Zhang, T. (2015, January 27–29). A fault diagnosis approach using SVM with data dimension reduction by PCA and LDA method. Proceedings of the 2015 Chinese Automation Congress (CAC), Wuhan, China.
32. A New SVM Method for an Indirect Matrix Converter With Common-Mode Voltage Reduction10.1109/TII.2013.2255032 ↗