1. Imbalanced learning: Improving classification of diabetic neuropathy from magnetic resonance imaging
- Author
-
Dinesh Selvarajah, Paul A. Armitage, Kevin Teh, Solomon Tesfaye, and Iain D. Wilkinson
- Subjects
Male ,Support Vector Machine ,Decision Analysis ,Computer science ,02 engineering and technology ,computer.software_genre ,Diagnostic Radiology ,Machine Learning ,0302 clinical medicine ,Endocrinology ,Medical Conditions ,Diabetic Neuropathies ,0202 electrical engineering, electronic engineering, information engineering ,Medicine and Health Sciences ,Oversampling ,Multidisciplinary ,Applied Mathematics ,Simulation and Modeling ,Radiology and Imaging ,Software Engineering ,Magnetic Resonance Imaging ,Multilayer perceptron ,Physical Sciences ,Medicine ,Engineering and Technology ,020201 artificial intelligence & image processing ,Female ,Management Engineering ,Algorithms ,Research Article ,Computer and Information Sciences ,Imaging Techniques ,Endocrine Disorders ,Science ,Decision tree ,Neuroimaging ,Machine learning ,Research and Analysis Methods ,03 medical and health sciences ,Machine Learning Algorithms ,Artificial Intelligence ,Diagnostic Medicine ,Support Vector Machines ,Diabetes Mellitus ,Humans ,Preprocessing ,business.industry ,Decision Trees ,Biology and Life Sciences ,Decision Tree Learning ,Support vector machine ,ComputingMethodologies_PATTERNRECOGNITION ,Metabolic Disorders ,Artificial intelligence ,business ,Classifier (UML) ,computer ,030217 neurology & neurosurgery ,Mathematics ,Neuroscience - Abstract
One of the fundamental challenges when dealing with medical imaging datasets is class imbalance. Class imbalance happens where an instance in the class of interest is relatively low, when compared to the rest of the data. This study aims to apply oversampling strategies in an attempt to balance the classes and improve classification performance. We evaluated four different classifiers from k-nearest neighbors (k-NN), support vector machine (SVM), multilayer perceptron (MLP) and decision trees (DT) with 73 oversampling strategies. In this work, we used imbalanced learning oversampling techniques to improve classification in datasets that are distinctively sparser and clustered. This work reports the best oversampling and classifier combinations and concludes that the usage of oversampling methods always outperforms no oversampling strategies hence improving the classification results.
- Published
- 2020