In order to increase the segmentation impact of the Diabetic Foot Ulcer Challenge 2022 dataset, we train a selection of popular deep learning segmentation algorithms and improve training methods, such as adding Dice term to loss function, employing transfer learning and poly learning rate update strategy, etc., in this paper. Experiments show that our method is effective, we get a Dice score of 0.7045, which is better than the official baseline result of 0.6277. Moreover, we integrate the above segmentation models using four ensemble methods to evaluate segmentation performance, such as Averaging, Weighting, Voting, and Stacking. We observed that our proposed one-layer CNN stacking network exhibits superior segmentation performance (Dice score: 0.7142) compared to single CNN model and other three ensemble methods. Our performance surpasses the baseline result, placing us in the top 10 in the Diabetic Foot Ulcer Challenge 2022.
Among the many methods of deep semi-supervised learning (DSSL), the holistic method combines ideas from other methods, such as consistency regularization and pseudo-labeling, with great success. This method typically introduces a threshold to utilize unlabeled data. If the highest predictive value from unlabeled data exceeds the threshold, the associated class is designated as the data’s pseudo-label. However, current methods utilize fixed or dynamic thresholds, disregarding the varying learning difficulties across categories in unbalanced datasets. To overcome these issues, in this paper, we first designed Cumulative Effective Labeling (CEL) to reflect a particular class’s learning difficulty. This approach differs from previous methods because it uses effective pseudo-labels and ground truth, collectively influencing the model’s capacity to acquire category knowledge. In addition, based on CEL, we propose a simple but effective way to compute the threshold, Self-adaptive Dynamic Threshold (SDT). It requires a single hyperparameter to adjust to various scenarios, eliminating the necessity for a unique threshold modification approach for each case. SDT utilizes a clever mapping function that can solve the problem of differential learning difficulty of various categories in an unbalanced image dataset that adversely affects dynamic thresholding. Finally, we propose a deep semi-supervised method with SDT called FldtMatch. Through theoretical analysis and extensive experiments, we have fully proven that FldtMatch can overcome the negative impact of unbalanced data. Regardless of the choice of the backbone network, our method achieves the best results on multiple datasets. The maximum improvement of the macro F1-Score metric is about 5.6% in DFUC2021 and 2.2% in ISIC2018.