Spatial Aggregation of Holistically-Nested Networks for Automated Pancreas Segmentation
This work addresses the problem of accurate pancreas segmentation for medical image analysis, which is incremental as it improves upon existing methods with a specific gain.
The paper tackles the challenge of automated pancreas segmentation in medical images by proposing a holistic learning approach that integrates semantic mid-level cues via spatial aggregation, achieving a Dice Similarity Coefficient of 78.01% ± 8.2%, significantly outperforming the previous state-of-the-art of 71.8% ± 10.7%.
Accurate automatic organ segmentation is an important yet challenging problem for medical image analysis. The pancreas is an abdominal organ with very high anatomical variability. This inhibits traditional segmentation methods from achieving high accuracies, especially compared to other organs such as the liver, heart or kidneys. In this paper, we present a holistic learning approach that integrates semantic mid-level cues of deeply-learned organ interior and boundary maps via robust spatial aggregation using random forest. Our method generates boundary preserving pixel-wise class labels for pancreas segmentation. Quantitative evaluation is performed on CT scans of 82 patients in 4-fold cross-validation. We achieve a (mean $\pm$ std. dev.) Dice Similarity Coefficient of 78.01% $\pm$ 8.2% in testing which significantly outperforms the previous state-of-the-art approach of 71.8% $\pm$ 10.7% under the same evaluation criterion.