Enhancing Weakly Supervised Semantic Segmentation with Multi-label Contrastive Learning and LLM Features Guidance.

Histopathological whole-slide image (WSI) segmentation is essential for precise tissue characterization in medical diagnostics. However, traditional approaches require labor-intensive pixel-level annotations. To this end, we study weakly supervised semantic segmentation (WSSS) which uses patch-level classification labels, reducing annotation efforts significantly. However, the complexity of WSIs and the challenge of sparse classification labels hinder effective dense pixel predictions. Moreover, due to the multi-label nature of WSI, existingapproachesofsingle-labelcontrastivelearningdesignedfortherepresentationofsingle-category, neglecting the presence of other relevant categories and thus fail to adapt to WSI tasks. This paper presents a novel multilabel contrastive learning method for WSSS by incorporating class-specific embedding extraction with LLM features guidance. Specifically, we propose to obtain class-specific embeddings by utilizing classifier weights, followed by a dot-product-based attention fusion method that leverages LLM features to enrich their semantics, facilitating contrastive learning between different classes from single image. Besides, we propose a Robust Learning approach that leverages multi-layer features to evaluate the uncertainty of pseudo-labels, thereby mitigating the impact of noisy pseudo-labels on the learning process of segmentation. Extensive experiments have been conducted on two Histopathological image segmentation datasets, i.e. LUAD dataset and BCSS dataset, demonstrating the effectiveness of our methods with leading performance.

Dynamic decoding and dual synthetic data for automatic correction of grammar in low-resource scenario.

Grammar error correction systems are pivotal in the field of natural language processing (NLP), with a primary focus on identifying and correcting the grammatical integrity of written text. This is crucial for both language learning and formal communication. Recently, neural machine translation (NMT) has emerged as a promising approach in high demand. However, this approach faces significant challenges, particularly the scarcity of training data and the complexity of grammar error correction (GEC), especially for low-resource languages such as Indonesian. To address these challenges, we propose InSpelPoS, a confusion method that combines two synthetic data generation methods: the Inverted Spellchecker and Patterns+POS. Furthermore, we introduce an adapted seq2seq framework equipped with a dynamic decoding method and state-of-the-art Transformer-based neural language models to enhance the accuracy and efficiency of GEC. The dynamic decoding method is capable of navigating the complexities of GEC and correcting a wide range of errors, including contextual and grammatical errors. The proposed model leverages the contextual information of words and sentences to generate a corrected output. To assess the effectiveness of our proposed framework, we conducted experiments using synthetic data and compared its performance with existing GEC systems. The results demonstrate a significant improvement in the accuracy of Indonesian GEC compared to existing methods.

DFTNet: Dual-Path Feature Transfer Network for Weakly Supervised Medical Image Segmentation.

Medical image segmentation has long suffered from the problem of expensive labels. Acquiring pixel-level annotations is time-consuming, labor-intensive, and relies on extensive expert knowledge. Bounding box annotations, in contrast, are relatively easy to acquire. Thus, in this paper, we explore to segment images through a novel Dual-path Feature Transfer design with only bounding box annotations. Specifically, a Target-aware Reconstructor is proposed to extract target-related features by reconstructing the pixels within the bounding box through the channel and spatial attention module. Then, a sliding Feature Fusion and Transfer Module (FFTM) fuses the extracted features from Reconstructor and transfers them to guide the Segmentor for segmentation. Finally, we present the Confidence Ranking Loss (CRLoss) which dynamically assigns weights to the loss of each pixel based on the network's confidence. CRLoss mitigates the impact of inaccurate pseudo-labels on performance. Extensive experiments demonstrate that our proposed model achieves state-of-the-art performance on the Medical Segmentation Decathlon (MSD) Brain Tumour and PROMISE12 datasets.

DMCGNet: A Novel Network for Medical Image Segmentation With Dense Self-Mimic and Channel Grouping Mechanism.

Automatic Medical Image Segmentation (MIS) can assist doctors by reducing labor and providing a unified standard. Nowadays, approaches based on Deep Learning have become mainstream for MIS because of their ability of automatic feature extraction. However, due to the plain network design and targets variety in medical images, the semantic features can hardly be extracted adequately. In this work, we propose a novel Dense Self-Mimic and Channel Grouping based Network (DMCGNet) for MIS for better feature extraction. Specifically, we introduce a Pyramid Target-aware Dense Self Mimic (PTDSM) module, which is capable of exploring deeper and better feature representation with no parameter increase. Then, to utilize features efficiently, an effective Channel Split based Feature Fusion Module (CSFFM) is proposed for feature reuse, which strengthens the adaptation of multi-scale targets by utilizing the channel grouping mechanism. Finally, to train the proposed method adequately, Deep Supervision with Group Ensemble Learning (DSGEL) is equipped to the network. Extensive experiments demonstrate that our proposed model achieves state-of-the-art performance on 4 medical image segmentation datasets.

Antioxidant activity and in vitro digestibility of dialdehyde starches as influenced by their physical and structural properties.

Dialdehyde starches (DASs) with different aldehyde contents were prepared by periodate oxidation of corn starch, and their antioxidant activity and digestibility were studied and related to their structural characteristics, including morphology, relative crystallinity, average molecular weights, swelling power and solubility. The results revealed that DASs were effective antioxidants as revealed by the 2,2'-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity. A significant positive correlation (r>0.959, p<0.01) was found between the antioxidant effect and the aldehyde contents of DASs at all concentrations. The oxidation of starch increased the rapidly digestible starch (RDS) and resistant starch (RS) contents but reduced the amount of slowly digestible starch (SDS). Correlation analysis indicated that the amounts of RDS and RS were positively correlated to the aldehyde contents of DASs (r=0.973, p<0.01, and r=0.900, p<0.01), whilst the amount of SDS was negatively correlated to the aldehyde contents (r=-0.960, p<0.01). The structural characteristic also plays an important role for the antioxidant activity and digestibility of DASs.