Simultaneous MR knee image segmentation and bias field correction using deep learning and partial convolution

摘要

Intensity inhomogeneity is a great challenge for automated organ segmentation in magnetic resonance (MR)images. Many segmentation methods fail to deliver satisfactory results when the images are corrupted by a biasfield. Although inhomogeneity correction methods exist, they often fail to remove the bias field completely inknee MR images. We present a new iterative approach that simultaneously predicts the segmentation mask ofknee structures using a 3D U-net and estimates the bias field in 3D MR knee images using partial convolutionoperations. First, the test images run through a trained 3D U-net to generate a preliminary segmentation result,which is then fed to the partial convolution filter to create a preliminary estimation of the bias field using thesegmented bone mask. Finally, the estimated bias field is then used to produce bias field corrected imagesas the new inputs to the 3D U-net. Through this loop, the segmentation results and bias field correction areiteratively improved. The proposed method was evaluated on 20 proton-density (PD)-weighted knee MRI scanswith manually created segmentation ground truth using 10 fold cross-validation. In our preliminary experiments,the proposed methods outperformed conventional inhomogeneity-correction-plus-segmentation setup in terms ofboth segmentation accuracy and speed.