Hua Jin

Hua Jin

Auto-generated profile, based on publicly available information

Hua Jin

Department of Anesthesiology, The First People's Hospital of Yunnan Province

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 February 2023 in IEEE Transactions on Knowledge and Data Engineering
Verified
Journal Article
Closed access
QZQihuang Zhong
Qihuang Zhong
LDLiang Ding
Liang Ding
JLJuhua Liu
Juhua Liu
BDBo Du
Bo Du
HJHua Jin
Hua Jin
DTDacheng Tao
Dacheng Tao
Aspect-based sentiment analysis (ABSA) is a fine-grained task of sentiment analysis. To better comprehend long complicated sentences and obtain accurate aspect-specific information, linguistic and commonsense knowledge are generally required in this task. However, most current methods employ complicated and inefficient approaches to incorporate external knowledge, e.g., directly searching the graph nodes. Additionally, the complementarity between external knowledge and linguistic information has not been thoroughly studied. To this end, we propose a knowledge graph augmented network ( KGAN ), which aims to effectively incorporate external knowledge with explicitly syntactic and contextual information. In particular, KGAN captures the sentiment feature representations from multiple different perspectives, i.e. , context-, syntax- and knowledge-based. First, KGAN learns the contextual and syntactic representations in parallel to fully extract the semantic features. Then, KGAN integrates the knowledge graphs into the embedding space, based on which the aspect-specific knowledge representations are further obtained via an attention mechanism. Last, we propose a hierarchical fusion module to complement these multi-view representations in a local-to-global manner. Extensive experiments on five popular ABSA benchmarks demonstrate the effectiveness and robustness of our KGAN. Notably, with the help of the pretrained model of RoBERTa, KGAN achieves a new record of state-of-the-art performance among all datasets.

7 February 2022 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
WXWanxin Xiao
Wanxin Xiao
YZyafei zhang
yafei zhang
HWHongbin Wang
Hongbin Wang
FLFan Li
Fan Li
HJHua Jin
Hua Jin
Recently, infrared–visible image fusion has attracted more and more attention, and numerous excellent methods in this field have emerged. However, when the low-resolution images are being fused, most fusion results are of low resolution, limiting the practical application of the fusion results. Although some methods can simultaneously realize the fusion and super-resolution of low-resolution images, the improvement of fusion performance is limited due to the lack of guidance of high-resolution fusion results. To address this issue, we propose a heterogeneous knowledge distillation network (HKDnet) with multilayer attention embedding to jointly implement the fusion and super-resolution of infrared and visible images. Precisely, the proposed method consists of a high-resolution image fusion network (teacher network) and a low-resolution image fusion and super-resolution network (student network). The teacher network mainly fuses the high-resolution input images and guides the student network to obtain the ability of joint implementation of fusion and super-resolution. In order to make the student network pay more attention to the texture details of the visible input image, we designed a corner embedding attention mechanism. The mechanism integrates channel attention, position attention, and corner attention to highlight the visible image’s edge, texture, and structure. For the input infrared image, the dual-frequency attention (DFA) is constructed by mining the relationship of interlayer features to highlight the role of salient targets of the infrared image in the fusion result. The experimental results show that compared with the existing methods, the proposed method preserves the image information of both visible and infrared modalities, achieves sound visual effects, and displays accurate and natural texture details. The code of the proposed method can be available at https://github.com/firewaterfire/HKDnet .

14 July 2023 in IEEE/ACM Transactions on Audio, Speech, and Language Processing
Verified
Journal Article
Closed access
JLJuhua Liu
Juhua Liu
QZQihuang Zhong
Qihuang Zhong
LDLiang Ding
Liang Ding
HJHua Jin
Hua Jin
BDBo Du
Bo Du
DTDacheng Tao
Dacheng Tao
The goal of aspect-based sentiment analysis (ABSA) is to determine the sentiment polarity towards an aspect. Because of the expensive and limited amounts of labelled data, the pretraining strategy has become the de facto standard for ABSA. However, there always exists a severe domain shift between the pretraining and downstream ABSA datasets, which hinders effective knowledge transfer when directly fine-tuning, making the downstream task suboptimal. To mitigate this domain shift, we introduce a unified alignment pretraining framework into the vanilla pretrain-finetune pipeline, that has both instance- and knowledge-level alignments. Specifically, we first devise a novel coarse-to-fine retrieval sampling approach to select target domain-related instances from the large-scale pretraining dataset, thus aligning the instances between pretraining and the target domains ( First Stage ). Then, we introduce a knowledge guidance-based strategy to further bridge the domain gap at the knowledge level. In practice, we formulate the model pretrained on the sampled instances into a knowledge guidance model and a learner model. On the target dataset, we design an on-the-fly teacher-student joint fine-tuning approach to progressively transfer the knowledge from the knowledge guidance model to the learner model ( Second Stage ). Therefore, the learner model can maintain more domain-invariant knowledge when learning new knowledge from the target dataset. In the Third Stage, the learner model is finetuned to better adapt its learned knowledge to the target dataset. Extensive experiments and analyses on several ABSA benchmarks demonstrate the effectiveness and universality of our proposed pretraining framework.

3 November 2016 in Oncotarget
Verified
Journal Article
Open access
PDF file
THTao Hu
Tao Hu
YCYe-Fei Chang
Ye-Fei Chang
ZXZhangang Xiao
Zhangang Xiao
RMRui Mao
Rui Mao
JTJun Tong
Jun Tong
HJHua Jin
Hua Jin
MiR-1 inhibits progression of high-risk papillomavirus-associated human cervical cancer by targeting G6PD miR-1 inhibits progression of high-risk papillomavirus-associated human cervical cancer by targeting G6PD

1 October 2022 in Journal of Thoracic Disease
Verified
Journal Article
Open access
PDF file
YHYu Huang
Yu Huang
YBYun Bo
Yun Bo
YLYujin Li
Yujin Li
YZYanhua Zhao
Yanhua Zhao
XLXinnan Li
Xinnan Li
HJHua Jin
Hua Jin
Original Article The impact of tubeless anesthesia versus intubated anesthesia on cerebral oxygen saturation and postoperative cognitive function in patients undergoing video-assisted thoracoscopic surgery: a randomized trial Yu Huang, Yun Bo, Yujin Li, Yanhua Zhao, Xinnan Li, Dalin Chen, Na Deng, Jun Liu, Junjie Li, Mingzhen Zhu, Rongsheng Liu, Jun Peng, Hua Jin

1 October 2022 in Annals of Palliative Medicine
Verified
Journal Article
Open access
PDF file
JLJia Liu
Jia Liu
LLLungang Lin
Lungang Lin
XLXinnan Li
Xinnan Li
JPJun Peng
Jun Peng
DCDalin Chen
Dalin Chen
HJHua Jin
Hua Jin
Background: Pain management following cesarean section remains a challenge, with many puerpera suffering from severe acute postoperative pain. And for a second cesarean section the degree of uterine contraction pain is more severe and frequent than that of a primipara. This study investigated the effect of different doses of nalbuphine combined with sufentanil for postoperative analgesia in patients undergoing a second cesarean section. Methods: We prospectively recruited 168 women with a scarred uterus undergoing elective second cesarean section and they were randomly divided into 4 groups by random number extraction. A single intravenous injection of different doses of nalbuphine was given before the intravenous drip of oxytocin, and visual analogue scale (VAS) scores of uterine contraction pain were recorded 10 minutes before intravenous infusion of oxytocin (T1) and 10 minutes (T2), 30 minutes (T3), and 60 minutes (T4) after intravenous infusion of oxytocin. At 4, 8, 12, 24, and 48 hours after patient-controlled intravenous analgesia (PCIA), pain intensity was reassessed using the VAS score. Results: One hundred and sixty patients underwent elective second cesarean section in between December 2020 and May 2021 completed the study. The VAS scores of uterine contractions at T1 and T4 were 3 (1.0), while the VAS scores at T2 and T3 were 7 (1.0), 6 (1.0), 5 (1.0), 5 (1.0) and 8 (1.0), 5 (2.0), 3 (1.0), 3 (0.75). The VAS scores at 12 hours after surgery of nalbuphine10mg and sufentanil (NS1), nalbuphine 10 mg and sufentanil 20 mg (NS2) and nalbuphine 30 mg and sufentanil 20 mg (NS3) were lower than sufentanil (S) group (P<0.001). Compared with the S group, total amount of sufentanil and PCIA compression numbers in the NS1, NS2, and NS3 groups at 4–8 and 8–12 hours after surgery decreased (P<0.001), with a more significant decrease in the NS2 and NS3 groups than in the NS1 group (P<0.001). The NS3 group had a significantly higher incidence of dizziness and sleepiness (P=0.02, P=0.001). Compared with the NS2 and NS3 groups, the incidence of respiratory depression in the S group was significantly higher (P=0.001). Conclusions: A single intravenous injection of nalbuphine 20 mg 10 minutes before the infusion of oxytocin combined with sufentanil 2 µg/kg could be safely used for postoperative analgesia in patients undergoing a second cesarean section and could effectively inhibit uterine contractions induced by oxytocin and reduce adverse reactions. Trial Registration: Clinical Trial Registry ChiCTR2100042382.