One-Class Classification for Highly Imbalanced Medical Image Data

摘要

Computer-aided diagnosis plays an important role in clinical image diagnosis. Current clinical image classificationtasks usually focus on binary classification, which need to collect samples for both the positive and negative classesin order to train a binary classifier. However, in many clinical scenarios, there may have many more samples in oneclass than in the other class, which results in the problem of data imbalance. Data imbalance is a severe problem thatcan substantially influence the performance of binary-class machine learning models. To address this issue, oneclassclassification, which focuses on learning features from the samples of one given class, has been proposed. Inthis work, we assess the one-class support vector machine (OCSVM) to solve the classification tasks on two highlyimbalanced datasets, namely, space-occupying kidney lesions (including renal cell carcinoma and benign) data andbreast cancer distant metastasis/non-metastasis imaging data. Experimental results show that the OCSVM exhibitspromising performance compared to binary-class and other one-class classification methods.