Bootstrapping Semi-supervised Medical Image Segmentation with Anatomical-Aware Contrastive Distillation.

Contrastive learning has shown great promise over annotation scarcity problems in the context of medical image segmentation. Existing approaches typically assume a balanced class distribution for both labeled and unlabeled medical images. However, medical image data in reality is commonly imbalanced (i.e., multi-class label imbalance), which naturally yields blurry contours and usually incorrectly labels rare objects. Moreover, it remains unclear whether all negative samples are equally negative. In this work, we present ACTION, an Anatomical-aware ConTrastive dIstillatiON framework, for semi-supervised medical image segmentation. Specifically, we first develop an iterative contrastive distillation algorithm by softly labeling the negatives rather than binary supervision between positive and negative pairs. We also capture more semantically similar features from the randomly chosen negative set compared to the positives to enforce the diversity of the sampled data. Second, we raise a more important question: Can we really handle imbalanced samples to yield better performance? Hence, the key innovation in ACTION is to learn global semantic relationship across the entire dataset and local anatomical features among the neighbouring pixels with minimal additional memory footprint. During the training, we introduce anatomical contrast by actively sampling a sparse set of hard negative pixels, which can generate smoother segmentation boundaries and more accurate predictions. Extensive experiments across two benchmark datasets and different unlabeled settings show that ACTION significantly outperforms the current state-of-the-art semi-supervised methods.

Implicit Anatomical Rendering for Medical Image Segmentation with Stochastic Experts.

Integrating high-level semantically correlated contents and low-level anatomical features is of central importance in medical image segmentation. Towards this end, recent deep learning-based medical segmentation methods have shown great promise in better modeling such information. However, convolution operators for medical segmentation typically operate on regular grids, which inherently blur the high-frequency regions, i.e., boundary regions. In this work, we propose MORSE, a generic implicit neural rendering framework designed at an anatomical level to assist learning in medical image segmentation. Our method is motivated by the fact that implicit neural representation has been shown to be more effective in fitting complex signals and solving computer graphics problems than discrete grid-based representation. The core of our approach is to formulate medical image segmentation as a rendering problem in an end-to-end manner. Specifically, we continuously align the coarse segmentation prediction with the ambiguous coordinate-based point representations and aggregate these features to adaptively refine the boundary region. To parallelly optimize multi-scale pixel-level features, we leverage the idea from Mixture-of-Expert (MoE) to design and train our MORSE with a stochastic gating mechanism. Our experiments demonstrate that MORSE can work well with different medical segmentation backbones, consistently achieving competitive performance improvements in both 2D and 3D supervised medical segmentation methods. We also theoretically analyze the superiority of MORSE.

ACTION++: Improving Semi-supervised Medical Image Segmentation with Adaptive Anatomical Contrast.

Medical data often exhibits long-tail distributions with heavy class imbalance, which naturally leads to difficulty in classifying the minority classes (i.e., boundary regions or rare objects). Recent work has significantly improved semi-supervised medical image segmentation in long-tailed scenarios by equipping them with unsupervised contrastive criteria. However, it remains unclear how well they will perform in the labeled portion of data where class distribution is also highly imbalanced. In this work, we present ACTION++, an improved contrastive learning framework with adaptive anatomical contrast for semi-supervised medical segmentation. Specifically, we propose an adaptive supervised contrastive loss, where we first compute the optimal locations of class centers uniformly distributed on the embedding space (i.e., off-line), and then perform online contrastive matching training by encouraging different class features to adaptively match these distinct and uniformly distributed class centers. Moreover, we argue that blindly adopting a constant temperature τ in the contrastive loss on long-tailed medical data is not optimal, and propose to use a dynamic τ via a simple cosine schedule to yield better separation between majority and minority classes. Empirically, we evaluate ACTION++ on ACDC and LA benchmarks and show that it achieves state-of-the-art across two semi-supervised settings. Theoretically, we analyze the performance of adaptive anatomical contrast and confirm its superiority in label efficiency.

Rethinking Semi-Supervised Medical Image Segmentation: A Variance-Reduction Perspective.

For medical image segmentation, contrastive learning is the dominant practice to improve the quality of visual representations by contrasting semantically similar and dissimilar pairs of samples. This is enabled by the observation that without accessing ground truth labels, negative examples with truly dissimilar anatomical features, if sampled, can significantly improve the performance. In reality, however, these samples may come from similar anatomical regions and the models may struggle to distinguish the minority tail-class samples, making the tail classes more prone to misclassification, both of which typically lead to model collapse. In this paper, we propose ARCO, a semi-supervised contrastive learning (CL) framework with stratified group theory for medical image segmentation. In particular, we first propose building ARCO through the concept of variance-reduced estimation and show that certain variance-reduction techniques are particularly beneficial in pixel/voxel-level segmentation tasks with extremely limited labels. Furthermore, we theoretically prove these sampling techniques are universal in variance reduction. Finally, we experimentally validate our approaches on eight benchmarks, i.e., five 2D/3D medical and three semantic segmentation datasets, with different label settings, and our methods consistently outperform state-of-the-art semi-supervised methods. Additionally, we augment the CL frameworks with these sampling techniques and demonstrate significant gains over previous methods. We believe our work is an important step towards semi-supervised medical image segmentation by quantifying the limitation of current self-supervision objectives for accomplishing such challenging safety-critical tasks.

Converting Waste Polyethylene into ZnCCo3 and ZnCNi3 by a One-Step Thermal Reduction Process.

Plastic products have brought us great convenience in our daily life and work. But in the meantime, waste plastics have become solid pollutants in the environment due to its poor degradability. The resource utilization of waste plastic can decrease environmental pollution. Here, a thermal reduction method for the conversion of waste polyethylene to ZnCCo3 and ZnCNi3 in a stainless-steel autoclave under mild conditions has been reported. X-ray powder diffraction patterns indicate that the obtained samples are anti-perovskite-structured ternary carbides (ZnCCo3 and ZnCNi3) with good crystallinity. Moreover, the formation mechanism of ternary carbides has been briefly discussed. This method can be developed into an effective method for disposal of other waste plastics.

Facile Synthesis of Mo2C Nanoparticles from Waste Polyvinyl Chloride.

The resource utilization of waste plastic can not only control environmental pollution but can also ease up the problems of lack of energy resources. In this study, molybdenum carbide (Mo2C) nanoparticles have been synthesized by utilizing waste polyvinyl chloride as a carbon source in a stainless-steel autoclave at 600 °C. X-ray diffraction pattern indicates that the product is orthorhombic phase Mo2C. Electron microscopy photographs show that the obtained Mo2C product consisted of crystalline nanoparticles with an average size of 50 nm. The possible formation mechanisms of Mo2C have been also briefly discussed on the basis of the structures of the products synthesized with different reaction times. The effects of reaction temperature on the crystallinity and microstructure of the obtained products have been investigated. The results show that higher reaction temperature promotes the formation of Mo2C with high crystallinity.

2D molybdenum nitride nanosheets as anode materials for improved lithium storage.

Two-dimensional (2D) molybdenum nitride (MoN) nanosheets are promising anode materials for improved lithium-ion batteries. However, the reported synthesis methods of MoN generally rely on high-temperature and complex procedures with low cost efficiency. Herein, we report a facile one-pot synthesis of 2D MoN nanosheets at a low temperature of 400 °C via a solid-state reaction of molybdenum disulfide, sulfur and sodium amide in an autoclave. When employed as the anode material for lithium ion batteries, the as-developed MoN electrode exhibits outstanding cyclability with a high capacity retention of 898 mA h g-1 over 400 cycles at a current rate of 200 mA g-1 as well as a superb rate capability with a capacity of 505 mA h g-1 at a high rate up to 2 A g-1. The excellent lithium storage performance of the MoN electrode is attributed to its advantageously high conductivity and unique 2D nanostructure.