Automatic Cardiac Disease Assessment on cine-MRI via Time-Series Segmentation and Domain Specific Features
share
Cardiac magnetic resonance imaging improves on diagnosis of cardiovascular diseases by providing images at high spatiotemporal resolution. Manual evaluation of these time-series, however, is expensive and prone to biased and non-reproducible outcomes. In this paper, we present a method that addresses named limitations by integrating seg- mentation and disease classification into a fully automatic processing pipeline. We use a UNet inspired architecture for segmentation of car- diac structures such as the left and right ventricular cavity (LVC, RVC) and the left ventricular myocardium (LVM) on each time instance of the cardiac cycle. For the classification task, information is extracted from the segmented time-series in form of comprehensive features handcrafted to reflect diagnostic clinical procedures. Based on these features we train an ensemble of heavily regularized multilayer perceptrons (MLP) and a random forest classifier to predict the pathologic target class. We evalu- ated our method on the ACDC training dataset (4 pathology groups, 1 healthy group, 20 patients per group). We achieved dice scores of 0.930 (LVC), 0.879 (RVC) and 0.873 (LVM) and a classification accuracy of 94 Our results underpin the potential of machine learning methods for accurate, fast and reproducible segmenta- tion and computer-assisted diagnosis (CAD).
READ FULL TEXT