Gang Yu

Gang Yu


Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
1 September 2020 in The Journal of Pediatrics
Verified
Journal Article
Closed access
HLHaomin Li
Haomin Li
GYGang Yu
Gang Yu
HDHuilong Duan
Huilong Duan
JFJunfen Fu
Junfen Fu
QSQiang Shu
Qiang Shu

8 March 2022 in Frontiers in Pediatrics
Verified
Journal Article
Open access
WXWeize Xu
Weize Xu
LSLiqi Shu
Liqi Shu
PGPing Gong
Ping Gong
CHChencui Huang
Chencui Huang
JXJingxu Xu
Jingxu Xu
GYGang Yu
Gang Yu
Background: Developmental dysplasia of the hip (DDH) is a common orthopedic disease in children. In clinical surgery, it is essential to quickly and accurately locate the exact position of the lesion, and there are still some controversies relating to DDH status. We adopt artificial intelligence (AI) to solve the above problems.Methods: In this paper, automatic DDH measurements and classifications were achieved using a three-stage pipeline. In the first stage, we used Mask-RCNN to detect the local features of the image and segment the bony pelvis, including the ilium, pubis, ischium, and femoral heads. For the second stage, local image patches focused on semantically related areas for DDH landmarks were extracted by high-resolution network (HRNet). In the third stage, some radiographic results are obtained. In the above process, we used 1,265 patient x-ray samples as the training set and 133 samples from two other medical institutions as the verification set. The results of AI were compared with three orthopedic surgeons for reliability and time consumption.Results: AI-aided diagnostic system's Tönnis and International Hip Dysplasia Institute (IHDI) classification accuracies for both hips ranged from 0.86 to 0.95. The measurements of numerical indices showed that there was no statistically significant difference between surgeons and AI. Tönnis and IHDI indicators were similar across the AI system, intermediate surgeon, and junior surgeon. Among some objective interpretation indicators, such as acetabular index and CE angle, there were good stability and consistency among the four observers. Intraclass consistency of acetabular index and CE angle among surgeons was 0.79–0.98, while AI was 1.00. The measurement time required by AI was significantly less than that of the doctors.Conclusion: The AI-aided diagnosis system can quickly and automatically measure important parameters and improve the quality of clinical diagnosis and screening referral process with a convenient and efficient way.

15 February 2023 in Chemical Research in Toxicology
Verified
Journal Article
Open access
JHJian Huang
Jian Huang
JSJuan Su
Juan Su
ZHZhenyu Hou
Zhenyu Hou
JLJing Li
Jing Li
ZLZheming Li
Zheming Li
GYGang Yu
Gang Yu
Two-dimensional (2D) nanomaterials have been widely used in biomedical applications because of their biocompatibility. Considering the high risk of exposure of the circulatory system to Ti3C2Tx, we studied the cytocompatibility of Ti3C2Tx MXene with red blood cells (RBCs) and human umbilical vein endothelial cells (HUVECs) and showed that Ti3C2Tx had excellent compatibility with the two cell lines. Ti3C2Tx at a concentration as high as 200 μg/mL caused a negligible percent hemolysis of 0.8%. By contrast, at the same treatment concentration, graphene oxide (GO) caused a high percent hemolysis of 50.8%. Scanning electron microscopy revealed that RBC structures remained intact in the Ti3C2Tx treatment group, whereas those in the GO group completely deformed, sunk, and shrunk, which resulted in the release of cell contents. This difference can be largely ascribed to the distinct surficial properties of the two nanosheets. In specific, the fully covered surface-terminating -O and -OH groups leading to Ti3C2Tx had a very hydrophilic surface, thereby hindering its penetration into the highly hydrophobic interior of the cell membrane. However, the strong direct van der Waals attractions coordinated with hydrophobic interactions between the unoxidized regions of GO and the lipid hydrophobic tails can still damage the integrity of the cell membranes. In addition, the sharp and keen-edged corners of GO may also facilitate its relatively strong cell membrane damage effects than Ti3C2Tx. Thus, the excellent cell membrane compatibility of Ti3C2Tx nanosheets and their ultraweak capacity to provoke excessive ROS generation endowed them with much better compatibility with HUVECs than GO nanosheets. These results indicate that Ti3C2Tx has much better cytocompatibility than GO and provide a valuable reference for the future biomedical applications of Ti3C2Tx.

5 February 2025 in Alexandria Engineering Journal
Verified
Journal Article
Open access
JHJian Huang
Jian Huang
WDWen Ding
Wen Ding
TZTiancheng Zhong
Tiancheng Zhong
GYGang Yu
Gang Yu

8 January 2025 in Journal of Medical Internet Research
Verified
Journal Article
Open access
XLXiadong Li
Xiadong Li
QSQiang Shu
Qiang Shu
CKCanhong Kong
Canhong Kong
JWJinhu Wang
Jinhu Wang
GLGang Li
Gang Li
GYGang Yu
Gang Yu
Background: Accurate classification of patient complaints is crucial for enhancing patient satisfaction management in health care settings. Traditional manual methods for categorizing complaints often lack efficiency and precision. Thus, there is a growing demand for advanced and automated approaches to streamline the classification process. Objective: This study aimed to develop and validate an intelligent system for automatically classifying patient complaints using machine learning (ML) and natural language processing (NLP) techniques. Methods: An ML-based NLP technology was proposed to extract frequently occurring dissatisfactory words related to departments, staff, and key treatment procedures. A dataset containing 1465 complaint records from 2019 to 2023 was used for training and validation, with an additional 376 complaints from Hangzhou Cancer Hospital serving as an external test set. Complaints were categorized into 4 types—communication problems, diagnosis and treatment issues, management problems, and sense of responsibility concerns. The imbalanced data were balanced using the Synthetic Minority Oversampling Technique (SMOTE) algorithm to ensure equal representation across all categories. A total of 3 ML algorithms (Multifactor Logistic Regression, Multinomial Naive Bayes, and Support Vector Machines [SVM]) were used for model training and validation. The best-performing model was tested using a 5-fold cross-validation on external data. Results: The original dataset consisted of 719, 376, 260, and 86 records for communication problems, diagnosis and treatment issues, management problems, and sense of responsibility concerns, respectively. The Multifactor Logistic Regression and SVM models achieved weighted average accuracies of 0.89 and 0.93 in the training set, and 0.83 and 0.87 in the internal test set, respectively. Ngram-level term frequency–inverse document frequency did not significantly improve classification performance, with only a marginal 1% increase in precision, recall, and F1-score when implementing Ngram-level term frequency–inverse document frequency (n=2) from 0.91 to 0.92. The SVM algorithm performed best in prediction, achieving an average accuracy of 0.91 on the external test set with a 95% CI of 0.87-0.97. Conclusions: The NLP-driven SVM algorithm demonstrates effective classification performance in automatically categorizing patient complaint texts. It showed superior performance in both internal and external test sets for communication and management problems. However, caution is advised when using it for classifying sense of responsibility complaints. This approach holds promises for implementation in medical institutions with high complaint volumes and limited resources for addressing patient feedback.

10 February 2020 in JMIR Medical Informatics
Verified
Journal Article
Open access
JLJun Liang
Jun Liang
YLYing Li
Ying Li
ZZZhongan Zhang
Zhongan Zhang
DSDongxia Shen
Dongxia Shen
JXJie Xu
Jie Xu
Jianbo LeiJianbo Lei
Jianbo Lei
Background: To achieve universal access to medical resources, China introduced its second health care reform in 2010, with health information technologies (HIT) as an important technical support point. Objective: This study is the first attempt to explore the unique contributions and characteristics of HIT development in Chinese hospitals from the three major aspects of hospital HIT—human resources, funding, and materials—in an all-around, multi-angled, and time-longitudinal manner, so as to serve as a reference for decision makers in China and the rest of the world when formulating HIT development strategies. Methods: A longitudinal research method is used to analyze the results of the CHIMA Annual Survey of Hospital Information System in China carried out by a Chinese national industrial association, CHIMA, from 2007 to 2018. The development characteristics of human resources, funding, and materials of HIT in China for the past 12 years are summarized. The Bass model is used to fit and predict the popularization trend of EMR in Chinese hospitals from 2007 to 2020. Results: From 2007 to 2018, the CHIMA Annual Survey interviewed 10,954 hospital CIOs across 32 administrative regions in Mainland China. Compared with 2007, as of 2018, in terms of human resources, the average full time equivalent (FTE) count in each hospital’s IT center is still lower than the average level of US counterparts in 2014 (9.66 FTEs vs. 34 FTEs). The proportion of CIOs with a master’s degree or above was 25.61%, showing an increase of 18.51%, among which those with computer-related backgrounds accounted for 64.75%, however, those with a medical informatics background only accounted for 3.67%. In terms of funding, the sampled hospitals’ annual HIT investment increased from ¥957,700 (US $136,874) to ¥6.376 million (US $911,261), and the average investment per bed increased from ¥4,600 (US $658) to ¥8,100 (US $1158). In terms of information system construction, as of 2018, the average EMR implementation rate of the sampled hospitals exceeded the average level of their US counterparts in 2015 and their German counterparts in 2017 (85.26% vs. 83.8% vs. 68.4%, respectively). The results of the Bass prediction model show that Chinese hospitals will likely reach an adoption rate of 91.4% by 2020 (R2=0.95). Conclusions: In more than 10 years, based on this top-down approach, China’s medical care industry has accepted government instructions and implemented the unified model planned by administrative intervention. With only about one-fifth of the required funding, and about one-fourth of the required human resources per hospital as compared to the US HITECH project, China’s EMR coverage in 2018 exceeded the average level of its US counterparts in 2015 and German counterparts in 2017. This experience deserves further study and analysis by other countries.