An Embarrassingly Simple Consistency Regularization Method for Semi-Supervised Medical Image Segmentation

Abstract

The scarcity of pixel-level annotation is a prevalent problem in medical image segmentation tasks. In this paper, we introduce a novel regularization strategy involving interpolationbased mixing for semi-supervised medical image segmentation. The proposed method is a new consistency regularization strategy that encourages segmentation of interpolation of two unlabelled data to be consistent with the interpolation of segmentation maps of those data. This method represents a specific type of data-adaptive regularization paradigm which aids to minimize the overfitting of labelled data under high confidence values. The proposed method is advantageous over adversarial and generative models as it requires no additional computation. Upon evaluation on two publicly available MRI datasets: ACDC and MMWHS, experimental results demonstrate the superiority of the proposed method in comparison to existing semi-supervised models. Code is available at: https://github.com/hritam-98/ICT-MedSeg

Multi-organ segmentation: a progressive exploration of learning paradigms under scarce annotation

“…This method effectively improves segmentation results under a small number of supervised data, but yields little improvement compared to fully-supervised methods when the number is large. Similarly, Basak et al conducted effective semi-supervised segmentation by interpolation transform consistency using Mixup for data transformation , which extends the training set by mixing diverse images (Basak et al 2022).…”

Learning with Limited Annotations: A Survey on Deep Semi-Supervised Learning for Medical Image Segmentation

“…Huang et al [34] add cutout content loss and slice misalignment as input perturbations. Another common consistency is mix-up consistency [37], [38], [39], which encourages the segmentation of interpolation of two data to be consistent with the interpolation of segmentation results of those data. Apart from disturbances on inputs, there are also many studies focusing on disturbances at feature map level.…”