Zhongyi Zhou

Zhongyi Zhou

Auto-generated profile, based on publicly available information

Zhongyi Zhou

Gansu Provincial Key Laboratory of Wearable Computing School of Information Science and Engineering, Lanzhou University, Lanzhou, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
28 June 2023 in IEEE Transactions on Affective Computing
Verified
Journal Article
Closed access
XZXiaowei Zhang
Xiaowei Zhang
XWXiangyu Wei
Xiangyu Wei
ZZZhongyi Zhou
Zhongyi Zhou
QZQiqi Zhao
Qiqi Zhao
SZSipo Zhang
Sipo Zhang
BHBin Hu
Bin Hu
Stress has been identified as one of major causes of health issues. To detect the stress levels with higher accuracy, fusion of multimodal physiological signals is a promising technique. However, there is an asynchrony between physiological signals observed from different perspectives. Exploring the temporal alignment relationship between modalities is helpful to improve the quality of multimodal fusion. This paper proposes an end-to-end multimodal stress detection model based on Bidirectional Cross- and Self-modal Attention (BCSA) mechanism. Specifically, we first construct different feature extractors based on the characteristics of Blood Volume Pulse (BVP) and Electrodermal Activity (EDA) to complete automated temporal feature extraction. Secondly, cross-modal attention is used to seek the alignment relationship between the two modalities and fully fuse cross-modal information. The self-modal attention is used to attenuate noise and redundant information, highlight important information and obtain salient stress representations. Finally, the stress representations of the two modalities are processed separately, and the mean square error (MSE) is used to narrow the gap between them. Experimental results on the UBFC-Phys dataset and WESAD dataset show that the proposed model can effectively improve the accuracy of stress recognition, and outperforms several state-of-the-art methods.

1 January 2019 in Therapeutic Advances in Gastroenterology
Verified
Journal Article
Open access
PDF file
YLYuqiang Li
Yuqiang Li
LZLilan Zhao
Lilan Zhao
CGCenap Güngör
Cenap Güngör
FTFengbo Tan
Fengbo Tan
ZZZhongyi Zhou
Zhongyi Zhou
WLWenxue Liu
Wenxue Liu
There is no conclusion about the most important contributor to the upswing of locally advanced colorectal cancer (LACRC) survival. Data from the Surveillance, Epidemiology, and End Results (SEER) database was extracted to identify colorectal adenocarcinoma cancer patients at stage II and III diagnosed in the two periods 1989–1990 and 2009–2010. The statistical methods included Pearson’s chi-squared test, log-rank test, Cox regression model and propensity score matching. The Cox regression model showed that hazard ratio (HR) of non-surgery dropped from 11.529 to 3.469 in right colon cancer (RCC), 5.214 to 2.652 in left colon cancer (LCC) and 3.275 to 3.269 in rectal cancer (RC) from 1989–1990 to 2009–2010. The 95% confidence intervals (CIs) for surgical resection in 2009–2010 were narrower than those in 1989–1990. HR became greater in LACRC without chemotherapy (from 1.337 to 1.779 in RCC, 1.269 to 2.017 in LCC, 1.317 to 1.811 in RC). There was no overlapping about the 95% CI of chemotherapy between the two groups. The progress of surgery was not linked to the improvement of overall survival (OS) of RCC (p = 0.303) and RC (p = 0.660). Chemotherapy had a significant association with OS of all colorectal cancer (CRC) patients (p = 0.017 in RCC; p = 0.006 in LCC; p = 0.001 in RC). Advancements in chemotherapy regimen were the main contributor to the upswing of CRC survival. The improvements in surgery had a limited effect on improvements in CRC survival.

6 September 2022 in arXiv
Verified
Preprint
Open access
PDF file
ZZZhongyi Zhou
Zhongyi Zhou
Humans are talented with the ability to perform diverse interactions in the teaching process. However, when humans want to teach AI, existing interactive systems only allow humans to perform repetitive labeling, causing an unsatisfactory teaching experience. My Ph.D. research studies Interactive Machine Teaching (IMT), an emerging field of HCI research that aims to enhance humans' teaching experience in the AI creation process. My research builds IMT systems that exploit and guide user interaction and shows that such in-depth integration of human interaction can benefit both AI models and user experience.