-

 

 

 




Optimization Online





 

Expert-Enhanced Machine Learning for Cardiac Arrhythmia Classification

Sebastian Sager(sager***at***ovgu.de)
Felix Bernhardt(felix.bernhardt***at***ovgu.de)
Florian Kehrle(f.kehrle***at***gmail.com)
Maximilian Merkert(maximilian.merkert***at***ovgu.de)
Andreas Potschka(andreas.potschka***at***iwr.uni-heidelberg.de)
Benjamin Meder(Benjamin.Meder***at***med.uni-heidelberg.de)
Hugo Katus(hugo.katus***at***med.uni-heidelberg.de)
Eberhard Scholz(eberhard.scholz***at***med.uni-heidelberg.de)

Abstract: Machine learning (ML) methodology has been suc- cessfully applied to many classification problems in medicine and beyond. Whereas the accuracy is often astonishing, the interpretability of the results has become an ubiquitous issue. In order to overcome this important but unsolved challenge, we propose to first reduce the complexity of the data, and then to combine the interpretability of expert systems with the deductive power of data driven ML. As a showcase we considered the arguably most difficult classification case of cardiac arrhythmias. Here the largest database with the gold standard (intracardiac measurements after invasive procedures) only contains 380 samples, yielding an additional challenge to ML. Still, our approach achieved an accuracy of 82.84%. The main advantage however is the interpretability of the classification results. Our features give insight into a possibly occurring multi-level atrioventricular blocking mechanism, which might improve treatment decisions and is thus an important step in the realization of personalized medicine. The idea to use mathematical modeling and optimization to generate new and clinically interpretable features for ML can be transferred to other cases of clinical decision support.

Keywords: Optimization, MINLP, Machine learning, Healthcare, Decision support, Combinatorial algorithms

Category 1: Applications -- Science and Engineering (Biomedical Applications )

Category 2: Combinatorial Optimization

Citation: submitted to IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS

Download: [PDF]

Entry Submitted: 10/10/2019
Entry Accepted: 10/10/2019
Entry Last Modified: 10/10/2019

Modify/Update this entry


  Visitors Authors More about us Links
  Subscribe, Unsubscribe
Digest Archive
Search, Browse the Repository

 

Submit
Update
Policies
Coordinator's Board
Classification Scheme
Credits
Give us feedback
Optimization Journals, Sites, Societies
Mathematical Optimization Society