Key result
A patch-based ensemble and transfer learning framework achieved a classification accuracy of 72.5% and Cohen's Kappa of 0.6309 for detecting motion artefacts in cardiac MRI.
Why the study?
Cardiac MRI is often degraded by motion artefacts from breath-hold difficulty, but time-consuming manual quality classification impedes end-to-end computer-aided diagnostic systems.
An ensemble and transfer learning framework can automatically classify motion artefacts in cardiac MRI with 72.5% accuracy, facilitating end-to-end computer-aided diagnosis.
Cardiac Magnetic Resonance Imaging (MRI) plays an important role in the analysis of cardiac function. However, the acquisition is often accompanied by motion artefacts because of the difficulty of breath-hold, especially for acute symptoms patients. Therefore, it is essential to assess the quality of cardiac MRI for further analysis. Time-consuming manual-based classification is not conducive to the construction of an end-to-end computer aided diagnostic system. To overcome this problem, an automatic cardiac MRI quality estimation framework using ensemble and transfer learning is proposed in this work. Multiple pre-trained models were initialised and fine-tuned on 2-dimensional image patches sampled from the training data. In the model inference process, decisions from these models are aggregated to make a final prediction. The framework has been evaluated on CMRxMotion grand challenge (MICCAI 2022) dataset which is small, multi-class, and imbalanced. It achieved a classification accuracy of 78.8% and 70.0% on the training set (5-fold cross-validation) and a validation set, respectively. The final trained model was also evaluated on an independent test set by the CMRxMotion organisers, which achieved the classification accuracy of 72.5% and Cohen's Kappa of 0.6309 (ranked top 1 in this grand challenge). Our code is available on Github: https://github.com/ruizhe-l/CMRxMotion.
No takes yet. Share an insight, caveat, or question.
Li et al. (2022) studied Motion artefacts in Cardiac MRI (n=40). Patch-based ensemble and transfer learning framework vs. Single classification models was evaluated on Classification accuracy on independent test set (Cohen's Kappa 0.6309). A patch-based ensemble and transfer learning framework achieved a classification accuracy of 72.5% and Cohen's Kappa of 0.6309 for detecting motion artefacts in cardiac MRI.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: