Observer variation-aware medical image segmentation by combining deep learning and surrogate-assisted genetic algorithms
This addresses the issue of heterogeneous training data due to observer variation in medical imaging, which is incremental as it builds on existing deep learning methods by explicitly modeling segmentation styles.
The paper tackles the problem of observer variation in medical image segmentation by training separate neural networks on automatically determined subgroups of data, achieving up to 23% improvement in Dice and surface Dice coefficients compared to a single network trained on all data.
There has recently been great progress in automatic segmentation of medical images with deep learning algorithms. In most works observer variation is acknowledged to be a problem as it makes training data heterogeneous but so far no attempts have been made to explicitly capture this variation. Here, we propose an approach capable of mimicking different styles of segmentation, which potentially can improve quality and clinical acceptance of automatic segmentation methods. In this work, instead of training one neural network on all available data, we train several neural networks on subgroups of data belonging to different segmentation variations separately. Because a priori it may be unclear what styles of segmentation exist in the data and because different styles do not necessarily map one-on-one to different observers, the subgroups should be automatically determined. We achieve this by searching for the best data partition with a genetic algorithm. Therefore, each network can learn a specific style of segmentation from grouped training data. We provide proof of principle results for open-sourced prostate segmentation MRI data with simulated observer variations. Our approach provides an improvement of up to 23% (depending on simulated variations) in terms of Dice and surface Dice coefficients compared to one network trained on all data.