A fully 3D multi-path convolutional neural network with feature fusion and feature weighting for automatic lesion identification in brain MRI images
This work addresses stroke lesion prediction from MRI for medical imaging applications, representing an incremental improvement over prior methods.
The paper tackles automatic lesion identification in brain MRI images by proposing a fully 3D multi-path convolutional neural network with feature fusion and weighting, achieving cross-validation accuracies of 60.5% on the ATLAS benchmark and 65% on multi-modal datasets, outperforming existing models like DeepMedic and 3D U-Net.
We propose a fully 3D multi-path convolutional network to predict stroke lesions from 3D brain MRI images. Our multi-path model has independent encoders for different modalities containing residual convolutional blocks, weighted multi-path feature fusion from different modalities, and weighted fusion modules to combine encoder and decoder features. Compared to existing 3D CNNs like DeepMedic, 3D U-Net, and AnatomyNet, our networks achieves the highest statistically significant cross-validation accuracy of 60.5% on the large ATLAS benchmark of 220 patients. We also test our model on multi-modal images from the Kessler Foundation and Medical College Wisconsin and achieve a statistically significant cross-validation accuracy of 65%, significantly outperforming the multi-modal 3D U-Net and DeepMedic. Overall our model offers a principled, extensible multi-path approach that outperforms multi-channel alternatives and achieves high Dice accuracies on existing benchmarks.