Joined Type Length Encoding for Nested Named Entity Recognition

In this article, we propose a new encoding scheme for named entity recognition (NER) called Joined Type-Length encoding (JoinedTL). Unlike most existing named entity encoding schemes, which focus on flat entities, JoinedTL can label nested named entities in a single sequence. JoinedTL uses a packed encoding to represent both type and span of a named entity, which not only results in less tagged tokens compared to existing encoding schemes, but also enables it to support nested NER. We evaluate the effectiveness of JoinedTL for nested NER on three nested NER datasets: GENIA in English, GermEval in German, and PerNest, our newly created nested NER dataset in Persian. We apply CharLSTM+WordLSTM+CRF, a three-layer sequence tagging model on three datasets encoded using JoinedTL and two existing nested NE encoding schemes, i.e., JoinedBIO and JoinedBILOU. Our experiment results show that CharLSTM+WordLSTM+CRF trained with JoinedTL encoded datasets can achieve competitive F1 scores as the ones trained with datasets encoded by two other encodings, but with 27%–48% less tagged tokens. To leverage the power of three different encodings, i.e., JoinedTL, JoinedBIO, and JoinedBILOU, we propose an encoding-based ensemble method for nested NER. Evaluation results show that the ensemble method achieves higher F1 scores on all datasets than the three models each trained using one of the three encodings. By using nested NE encodings including JoinedTL with CharLSTM+WordLSTM+CRF, we establish new state-of-the-art performance with an F1 score of 83.7 on PerNest, 74.9 on GENIA, and 70.5 on GermEval, surpassing two recent neural models specially designed for nested NER.

Magnetic resonance imaging in the evaluation of cervical foraminal stenosis: comparison of 3D T2 SPACE with sagittal oblique 2D T2 TSE

Objective The oblique orientation of the cervical neural foramina challenges the implementation of a short MRI protocol with concurrent excellent visualization of the spine. While sagittal oblique T2-weighted sequences permit good evaluation of the cervical neuroforamina, all segments may not be equally well depicted on a single sequence and conspicuity of foraminal stenosis may be limited. 3D T2-weighted sequences can be reformatted in arbitrary planes, including the sagittal oblique. We set out to compare 3D T2w SPACE sequences with sagittal oblique reformations and sagittal oblique 2D T2w TSE sequences for the evaluation of cervical foraminal visibility and stenosis. Materials and methods Sixty consecutive patients who underwent MRI of the cervical spine with sagittal oblique 2D T2w TSE and 3D T2w SPACE sequences were included. Image homogeneity of the sequences was evaluated. Imaging sets were assessed for structure visibility and foraminal stenosis by two independent readers. Results of the sequences were compared by Wilcoxon matched-pairs tests. Interreader agreement was evaluated by weighted κ. Results Visibility of most structures was rated good to excellent on both sequences (mean visibility scores ≥ 4.5 of 5), though neuroforaminal contents were better seen on sagittal oblique T2w TSE (mean scores 4.1–4.6 vs. 3.1–4.1 on 3D T2w SPACE, p < 0.01). Stenosis grades were comparable between sequences (mean 1.1–2.6 of 4), with slightly higher values for 3D T2w SPACE at some levels (difference ≤ 0.3 points). Conclusion 3D T2w SPACE is comparable with sagittal oblique 2D T2w TSE in the evaluation of cervical neural foramina.

Compaction pressure distribution and pressure uniformity of segmented rollers for automated fiber placement

For automated fiber placement onto molds with complex surfaces, uneven compaction pressure distribution limits tows number in a single sequence and affects layup quality. Compaction roller has a direct influence on the pressure distribution, but the relationship between the two has not been widely explored. In this paper, the segmented compaction roller is used, and a theoretical model of compaction pressure distribution for layup onto general surfaces is established by analyzing the contact between the roller and prepreg layers, followed by experimental validation. Based on the model, single-point pressure uniformity and whole-path pressure uniformity are proposed to quantitatively evaluate the pressure distribution. Furthermore, pressure distribution and pressure uniformity of segmented roller and common roller are compared, as well as the influence of the two pressure distribution on layup quality. The results show that the established model can predict pressure distribution and provide a basis for analyzing layup defects, and segmented rollers apply evener compaction pressure and help to improve layup quality.

Automatic Segmentation Algorithm of Magnetic Resonance Image in Diagnosis of Liver Cancer Patients under Deep Convolutional Neural Network

To study the influence of different sequences of magnetic resonance imaging (MRI) images on the segmentation of hepatocellular carcinoma (HCC) lesions, the U-Net was improved. Moreover, deep fusion network (DFN), data enhancement strategy, and random data (RD) strategy were introduced, and a multisequence MRI image segmentation algorithm based on DFN was proposed. The segmentation experiments of single-sequence MRI image and multisequence MRI image were designed, and the segmentation result of single-sequence MRI image was compared with those of convolutional neural network (FCN) algorithm. In addition, RD experiment and single-input experiment were also designed. It was found that the sensitivity (0.595 ± 0.145) and DSC (0.587 ± 0.113) obtained by improved U-Net were significantly higher than the sensitivity (0.405 ± 0.098) and DSC (0.468 ± 0.115, P < 0.05 ) obtained by U-Net. The sensitivity of multisequence MRI image segmentation algorithm based on DFN (0.779 ± 0.015) was significantly higher than that of FCN algorithm (0.604 ± 0.056, P < 0.05 ). The multisequence MRI image segmentation algorithm based on the DFN had higher indicators for liver cancer lesions than those of the improved U-Net. When RD was added, it not only increased the DSC of the single-sequence network enhanced by the hepatocyte-specific magnetic resonance contrast agent (Gd-EOB-DTPA) by 1% but also increased the DSC of the multisequence MRI image segmentation algorithm based on DFN by 7.6%. In short, the improved U-Net can significantly improve the recognition rate of small lesions in liver cancer patients. The addition of RD strategy improved the segmentation indicators of liver cancer lesions of the DFN and can fuse image features of multiple sequences, thereby improving the accuracy of lesion segmentation.