Abstract
We perform unsupervised analysis of image-derived shape and motion features extracted from 3,822 cardiac Magnetic resonance imaging (MRIs) of the UK Biobank. First, with a feature extraction method previously published based on deep learning models, we extract from each case 9 feature values characterizing both the cardiac shape and motion. Second, a feature selection is performed to remove highly correlated feature pairs. Third, clustering is carried out using a Gaussian mixture model on the selected features. After analysis, we identify 2 small clusters that probably correspond to 2 pathological categories. Further confirmation using a trained classification model and dimensionality reduction tools is carried out to support this finding. Moreover, we examine the differences between the other large clusters and compare our measures with the ground truth.
Highlights
In recent years, more and more data are made accessible for research in medical image analysis
There is only 1 point on the upper left corner of the plot representing a highly correlated pair. It corresponds to VLV,ES and EFLV, which are of Pearson correlation coefficient −0.80 and maximal information coefficient (MIC) 0.51
Analysis of the Resulting Clusters Among the 9 resulting clusters of the selected model, 2 are of small sizes. We find that they correspond to 2 pathological categories according to the definition given by the Automatic Cardiac Diagnosis Challenge (ACDC) (RVA and Dilated cardiomyopathy (DCM), respectively)
Summary
More and more data are made accessible for research in medical image analysis. Researchers are less constrained by the scarcity of data, which has been a prevailing challenge for a long time. One major challenge is how to make good use of unlabeled data [8, 9]. While there are more and more labeled data available, an important part of medical images is still unlabeled. This is understandable as it is in general expensive and tedious to diagnose and label cases by human experts. Methods that can extract useful information from unlabeled data are interesting and might potentially save a lot of time and effort
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.