Accurately identifying the pixels of small organs or lesions from magnetic resonance imaging (MRI) has a critical impact on clinical diagnosis. U-net is the most well-known and commonly used neural network for image segmentation. However, the small anatomical structures in medical images cannot be well recognised by U-net. This paper explores the performance of the U-net architectures in knee MRI segmentation to find a relative structure that can obtain high accuracies for both small and large anatomical structures. To maximise the utilities of U-net architecture, we apply three types of components, residual blocks, squeeze-and-excitation (SE) blocks, and dense blocks, to construct four variants of U-net, namely U-net variants. Among these variants, our experiments show that SE blocks can improve the segmentation accuracies of small labels. We adopt DeepLabv3plus architecture for 3D medical image segmentation by equipping SE blocks based on this discovery. The experimental results show that U-net with SE block achieves higher accuracy in parts of small anatomical structures. In contrast, DeepLabv3plus with SE block performs better on the average dice coefficient of small and large labels.

No abstract

Existing Sequence-to-Sequence (Seq2Seq) Neural Machine Translation (NMT) shows strong capability with High-Resource Languages (HRLs). However, this approach poses serious challenges when processing Low-Resource Languages (LRLs), because the model expression is limited by the training scale of parallel sentence pairs. This study utilizes adversary and transfer learning techniques to mitigate the lack of sentence pairs in LRL corpora. We propose a new Low resource, Adversarial, Cross-lingual (LAC) model for NMT. In terms of the adversary technique, LAC model consists of a generator and discriminator. The generator is a Seq2Seq model that produces the translations from source to target languages, while the discriminator measures the gap between machine and human translations. In addition, we introduce transfer learning on LAC model to help capture the features in rare resources because some languages share the same subject-verb-object grammatical structure. Rather than using the entire pretrained LAC model, we separately utilize the pretrained generator and discriminator. The pretrained discriminator exhibited better performance in all experiments. Experimental results demonstrate that the LAC model achieves higher Bilingual Evaluation Understudy (BLEU) scores and has good potential to augment LRL translations.

In some languages, Named Entity Recognition (NER) is severely hindered by complex linguistic structures, such as inflection, that will confuse the data-driven models when perceiving the word’s actual meaning. This work tries to alleviate these problems by introducing a novel neural network based on morphological and syntactic grammars. The experiments were performed in four Nordic languages, which have many grammar rules. The model was named the NorG network (Nor: Nordic Languages, G: Grammar). In addition to learning from the text content, the NorG network also learns from the word writing form, the POS tag, and dependency. The proposed neural network consists of a bidirectional Long Short-Term Memory (Bi-LSTM) layer to capture word-level grammars, while a bidirectional Graph Attention (Bi-GAT) layer is used to capture sentence-level grammars. Experimental results from four languages show that the grammar-assisted network significantly improves the results against baselines. We also investigate how the NorG network works on each grammar component by some exploratory experiments.

No abstract