FBN: Weakly Supervised Thyroid Nodule Segmentation Optimized by Online Foreground and Background.

OBJECTIVE: The main objective of the work described here was to train a semantic segmentation model using classification data for thyroid nodule ultrasound images to reduce the pressure of obtaining pixel-level labeled data sets. Furthermore, we improved the segmentation performance of the model by mining the image information to narrow the gap between weakly supervised semantic segmentation (WSSS) and fully supervised semantic segmentation. METHODS: Most WSSS methods use a class activation map (CAM) to generate segmentation results. However, the lack of supervision information makes it difficult for a CAM to highlight the object region completely. Therefore, we here propose a novel foreground and background pair (FB-Pair) representation method, which consists of high- and low-response regions highlighted by the original CAM-generated online in the original image. During training, the original CAM is revised using the CAM generated by the FB-Pair. In addition, we design a self-supervised learning pretext task based on FB-Pair, which requires the model to predict whether the pixels in FB-Pair are from the original image during training. After this task, the model will accurately distinguish between different categories of objects. RESULTS: Experiments on the thyroid nodule ultrasound image (TUI) data set revealed that our proposed method outperformed existing methods, with a 5.7% improvement in the mean intersection-over-union (mIoU) performance of segmentation compared with the second-best method and a reduction to 2.9% in the difference between the performance of benign and malignant nodules. CONCLUSION: Our method trains a well-performing segmentation model on ultrasound images of thyroid nodules using only classification data. In addition, we determined that CAM can take full advantage of the information in the images to highlight the target regions more accurately and thus improve the segmentation performance.

ATTransUNet: An enhanced hybrid transformer architecture for ultrasound and histopathology image segmentation.

Recently, researchers have introduced Transformer into medical image segmentation networks to encode long-range dependency, which makes up for the deficiencies of convolutional neural networks (CNNs) in global context modeling, and thus improves segmentation performance. However, in Transformer, due to the heavy computational burden of paired attention modeling between redundant visual tokens, the efficiency of Transformer needs to be further improved. Therefore, in this paper, we propose ATTransUNet, a Transformer enhanced hybrid architecture based on the adaptive token for ultrasound and histopathology image segmentation. In the encoding stage of the ATTransUNet, we introduced an Adaptive Token Extraction Module (ATEM), which can mine a few important visual tokens in the image for self-attention modeling, thus reducing the complexity of the model and improving the segmentation accuracy. In addition, in the decoding stage, we introduce a Selective Feature Reinforcement Module (SFRM) to reinforce the representation of and attention to key tissues or pathological features. The proposed ATTransUNet is evaluated on the basis of three medical image segmentation datasets. The results show that ATTransUNet achieves the best segmentation performance compared with the previous state-of-the-art models, and the proposed method is also competitive in terms of the network parameters and computation.