Jianping Yin’s research while affiliated with Dongguan University of Technology and other places

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.

Out-of-distribution (OOD) detection is a crucial problem in practice, especially, for the safe deployment of machine learning models in industrial settings. Previous work has used free energy as a score function and proposed a fine-tuning method that utilized OOD data in the training phase of the classification model, which achieves a higher performance on the OOD detection task compared with traditional methods. One key drawback, however, is that the loss function parameters are highly dependent on involved datasets, which means it cannot be dynamically adapted and implemented in others settings; in other words, the general ability of the energy score is considerably limited. In this work, our point of departure is to enlarge distinguishability between in-distribution features and OOD data. Consequently, we present a simple yet effective sparsity-regularized (SR) tuning framework for this purpose. Our framework has two types of workflows depending on if external OOD data is available, the complexity of the original training loss is sharply reduced by adopting this modification, meanwhile, the adapted ability and detection performance are enhanced. Also, we contribute a mini dataset as a light and efficient alternative of the previous large-scale one. In the experiments, we verify the effectiveness of our framework in a wide range of typical datasets along with common network architectures.

Models based on U-like structures have improved the performance of medical image segmentation. However, the single-layer decoder structure of U-Net is too "thin" to exploit enough information, resulting in large semantic differences between the encoder and decoder parts. Things get worse if the number of training sets of data is not sufficiently large, which is common in medical image processing tasks where annotated data are more difficult to obtain than other tasks. Based on this observation, we propose a novel U-Net model named MS-UNet for the medical image segmentation task in this study. Instead of the single-layer U-Net decoder structure used in Swin-UNet and TransUnet, we specifically design a multi-scale nested decoder based on the Swin Transformer for U-Net. The proposed multi-scale nested decoder structure allows the feature mapping between the decoder and encoder to be semantically closer, thus enabling the network to learn more detailed features. In addition, we propose a novel edge loss and a plug-and-play fine-tuning Denoising module, which not only effectively improves the segmentation performance of MS-UNet, but could also be applied to other models individually. Experimental results show that MS-UNet could effectively improve the network performance with more efficient feature learning capability and exhibit more advanced performance, especially in the extreme case with a small amount of training data, and the proposed Edge loss and Denoising module could significantly enhance the segmentation performance of MS-UNet.

The application of deep learning in the field of medical images has been gaining attention in recent years. However, due to the small number of medical images, the inability to generate them manually, and the high cost of annotation, it is difficult to train them. Masked Autoencoders (MAE), a recent hot new approach in the field of self-supervision, is based on Vision Transformer (ViT) architecture coding to improve the learning difficulty by adding masking to the cut non-overlapping image patches to obtain deeper image representations. In this paper, we employ three different small-scale pre-training datasets using MAE’s pre-training method to pre-train and select the medical segmentation task as its performance test task. We experimentally find that the new hybrid dataset pre-trained with the training set of the hybrid downstream task and other datasets has good performance, and the results are improved compared to those without pre-training, and better than those obtained with a single dataset, showing the potential of the self-supervised method hybrid pre-training for medical segmentation tasks.KeywordsMasked autoencodersHybrid pre-trainingMedical image segmentation