Yuqing Yang

Dr. Yuqing Yang


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
19 February 2024 in Journal of Applied Clinical Medical Physics
Verified
Journal Article
Open access
XWXin Wang
Xin Wang
SCSheng Cai
Sheng Cai
HWHongyan Wang
Hongyan Wang
JLJianchu Li
Jianchu Li
YYYuqing Yang
Yuqing Yang
Purpose: Diagnosing Renal artery stenosis (RAS) presents challenges. This research aimed to develop a deep learning model for the computer‐aided diagnosis of RAS, utilizing multimodal fusion technology based on ultrasound scanning images, spectral waveforms, and clinical information.Methods: A total of 1485 patients received renal artery ultrasonography from Peking Union Medical College Hospital were included and their color doppler sonography (CDS) images were classified according to anatomical site and left‐right orientation. The RAS diagnosis was modeled as a process involving feature extraction and multimodal fusion. Three deep learning (DL) models (ResNeSt, ResNet, and XCiT) were trained on a multimodal dataset consisted of CDS images, spectrum waveform images, and individual basic information. Predicted performance of different models were compared with senior physician and evaluated on a test dataset (N = 117 patients) with renal artery angiography results.Results: Sample sizes of training and validation datasets were 3292 and 169 respectively. On test data (N = 676 samples), predicted accuracies of three DL models were more than 80% and the ResNeSt achieved the accuracy 83.49% ± 0.45%, precision 81.89% ± 3.00%, and recall 76.97% ± 3.7%. There was no significant difference between the accuracy of ResNeSt and ResNet (82.84% ± 1.52%), and the ResNeSt was higher than the XCiT (80.71% ± 2.23%, p < 0.05). Compared to the gold standard, renal artery angiography, the accuracy of ResNest model was 78.25% ± 1.62%, which was inferior to the senior physician (90.09%). Besides, compared to the multimodal fusion model, the performance of single‐modal model on spectrum waveform images was relatively lower.Conclusion: The DL multimodal fusion model shows promising results in assisting RAS diagnosis.

27 May 2024 in BMC Medical Informatics and Decision Making
Verified
Journal Article
Open access
JZJun Zhou
Jun Zhou
XWXin Wang
Xin Wang
YLYiyao Li
Yiyao Li
YYYuqing Yang
Yuqing Yang
JSJuhong Shi
Juhong Shi
Background: Acute pulmonary thromboembolism (PTE) is a common cardiovascular disease and recognizing low prognosis risk patients with PTE accurately is significant for clinical treatment. This study evaluated the value of federated learning (FL) technology in PTE prognosis risk assessment while ensuring the security of clinical data. Methods: A retrospective dataset consisted of PTE patients from 12 hospitals were collected, and 19 physical indicators of patients were included to train the FL-based prognosis assessment model to predict the 30-day death event. Firstly, multiple machine learning methods based on FL were compared to choose the superior model. And then performance of models trained on the independent (IID) and non-independent identical distributed(Non-IID) datasets was calculated and they were tested further on Real-world data. Besides, the optimal model was compared with pulmonary embolism severity index (PESI), simplified PESI (sPESI), Peking Union Medical College Hospital (PUMCH). Results: The area under the receiver operating characteristic curve (AUC) of logistic regression(0.842) outperformed convolutional neural network (0.819) and multi layer perceptron (0.784). Under IID, AUC of model trained using FL(Fed) on the training, validation and test sets was 0.852 ± 0.002, 0.867 ± 0.012 and 0.829 ± 0.004. Under Real-world, AUC of Fed was 0.855 ± 0.005, 0.882 ± 0.003 and 0.835 ± 0.005. Under IID and Real-world, AUC of Fed surpassed centralization model(NonFed) (0.847 ± 0.001, 0.841 ± 0.001 and 0.811 ± 0.001). Under Non-IID, although AUC of Fed (0.846 ± 0.047) outperformed NonFed (0.841 ± 0.001) on validation set, it (0.821 ± 0.016 and 0.799 ± 0.031) slightly lagged behind NonFed (0.847 ± 0.001 and 0.811 ± 0.001) on the training and test sets. In practice, AUC of Fed (0.853, 0.884 and 0.842) outshone PESI (0.812, 0.789 and 0.791), sPESI (0.817, 0.770 and 0.786) and PUMCH(0.848, 0.814 and 0.832) on the training, validation and test sets. Additionally, Fed (0.842) exhibited higher AUC values across test sets compared to those trained directly on the clients (0.758, 0.801, 0.783, 0.741, 0.788). Conclusions: In this study, the FL based machine learning model demonstrated commendable efficacy on PTE prognostic risk prediction, rendering it well-suited for deployment in hospitals.

16 March 2022 in Respirology
Verified
Journal Article
Open access
YYYuqing Yang
Yuqing Yang
XWXin Wang
Xin Wang
YZYun‐Jian Zhang
Yun‐Jian Zhang
YCYan‐Fan Chen
Yan‐Fan Chen
LWLing Wang
Ling Wang
YJYing‐Qun Ji
Ying‐Qun Ji
Background and objective The pulmonary embolism severity index (PESI) and simplified PESI (sPESI) are recommended to recognize patients with acute pulmonary thromboembolism (PTE) with low prognosis risk, which is of great significance for treatment. This study aims to verify the influence of hypocalcaemia on the prognosis of patients with PTE and to establish a new prognosis assessment model. Methods This is an observational, multicentre study enrolling patients with PTE from February 2010 to June 2020 across 12 Chinese hospitals. Variables in PESI, serum calcium levels and patient survival status as of 5 July 2020 were collected. The area under the curve of the receiver operating characteristic curve, sensitivity, specificity and Youden index were used to evaluate model performance. Results In the cohort of 4196 patients with PTE, independent associations existed between hypocalcaemia and mid- and long-term mortalities (p <0.05). By including hypocalcaemia, the new 30-day death risk prediction rule, Peking Union Medical College Hospital rule (PUMCH rule), showed significantly higher specificity (0.622 [0.582, 0.661]; p <0.001) than the PESI (0.514 [0.473, 0.554]) and sPESI (0.484 [0.444, 0.525]) and similar sensitivity (0.963 [0.810, 0.999]; p = 0.161) with PESI (0.889 [0.708, 0.976]) and sPESI (0.963 [0.810, 0.999]) in the internal validation cohort. Well-performing predictive validity was also verified on a constructed external validation cohort. Conclusion Hypocalcaemia is independently associated with mid- and long-term PTE mortalities. The PUMCH rule showed significantly higher specificity than the PESI and sPESI and similar sensitivity, which may be used as a prognostic assessment tool for patients with acute PTE.

11 June 2025 in Journal of Medical Internet Research
Verified
Journal Article
Open access
RLRonghao Li
Ronghao Li
SMShuai Mao
Shuai Mao
CZCongmin Zhu
Congmin Zhu
YYYingliang Yang
Yingliang Yang
CTChunting Tan
Chunting Tan
YYYuqing Yang
Yuqing Yang
Background: The rapid advancements in natural language processing, particularly the development of large language models (LLMs), have opened new avenues for managing complex clinical text data. However, the inherent complexity and specificity of medical texts present significant challenges for the practical application of prompt engineering in diagnostic tasks. Objective: This paper explores LLMs with new prompt engineering technology to enhance model interpretability and improve the prediction performance of pulmonary disease based on a traditional deep learning model. Methods: A retrospective dataset including 2965 chest CT radiology reports was constructed. The reports were from 4 cohorts, namely, healthy individuals and patients with pulmonary tuberculosis, lung cancer, and pneumonia. Then, a novel prompt engineering strategy that integrates feature summarization (F-Sum), chain of thought (CoT) reasoning, and a hybrid retrieval-augmented generation (RAG) framework was proposed. A feature summarization approach, leveraging term frequency–inverse document frequency (TF-IDF) and K-means clustering, was used to extract and distill key radiological findings related to 3 diseases. Simultaneously, the hybrid RAG framework combined dense and sparse vector representations to enhance LLMs’ comprehension of disease-related text. In total, 3 state-of-the-art LLMs, GLM-4-Plus, GLM-4-air (Zhipu AI), and GPT-4o (OpenAI), were integrated with the prompt strategy to evaluate the efficiency in recognizing pneumonia, tuberculosis, and lung cancer. The traditional deep learning model, BERT (Bidirectional Encoder Representations from Transformers), was also compared to assess the superiority of LLMs. Finally, the proposed method was tested on an external validation dataset consisted of 343 chest computed tomography (CT) report from another hospital. Results: Compared with BERT-based prediction model and various other prompt engineering techniques, our method with GLM-4-Plus achieved the best performance on test dataset, attaining an F1-score of 0.89 and accuracy of 0.89. On the external validation dataset, F1-score (0.86) and accuracy (0.92) of the proposed method with GPT-4o were the highest. Compared to the popular strategy with manually selected typical samples (few-shot) and CoT designed by doctors (F1-score=0.83 and accuracy=0.83), the proposed method that summarized disease characteristics (F-Sum) based on LLM and automatically generated CoT performed better (F1-score=0.89 and accuracy=0.90). Although the BERT-based model got similar results on the test dataset (F1-score=0.85 and accuracy=0.88), its predictive performance significantly decreased on the external validation set (F1-score=0.48 and accuracy=0.78). Conclusions: These findings highlight the potential of LLMs to revolutionize pulmonary disease prediction, particularly in resource-constrained settings, by surpassing traditional models in both accuracy and flexibility. The proposed prompt engineering strategy not only improves predictive performance but also enhances the adaptability of LLMs in complex medical contexts, offering a promising tool for advancing disease diagnosis and clinical decision-making.

12 March 2026 in Advanced Functional Materials
Verified
Journal Article
Closed access
HYHe Yu
He Yu
CXChangyunkun Xiao
Changyunkun Xiao
TDTianyi Duan
Tianyi Duan
YYYuqing Yang
Yuqing Yang
YSYoungkee Shin
Youngkee Shin
Yang LiYang Li
Yang Li

26 March 2025 in Journal of Ultrasound in Medicine
Verified
Journal Article
Open access
XWXin Wang
Xin Wang
LMLiyuan Ma
Liyuan Ma
SPSiting Peng
Siting Peng
YYYuqing Yang
Yuqing Yang
JWJihao Wu
Jihao Wu
ZLZhenzhen Liu
Zhenzhen Liu
Objectives: Cesarean scar pregnancy (CSP) refers to a special type of pregnancy with a variable prognosis. We aimed to establish a prognostic classification system using ultrasound and clinical features to provide a reference for management strategies.Methods: Exactly 230 patients with CSP were included and categorized into three groups based on treatment and prognosis: Group A (favorable prognosis), Group B (moderate prognosis), and Group C (poor prognosis). A total of 26 ultrasound features and 8 clinical features were collected for further analysis. Machine learning and traditional scoring models were then constructed for Group A and Group C and integrated to predict CSP prognosis using the significant features.Results: In the univariate analysis, 26 variables were significantly correlated with Group C, while 21 variables were significantly correlated with Group A. For Group C, a linear scoring model was established using three key features: the criteria length of the implantation portion (IMPL) ≥2.43 cm, the height of the gestational sac or mass protruding above the uterine cavity line (GSUCL) ≥1.4 cm, and absent residual myometrial thickness (RMT), achieving an area under the curve (AUC) of 0.939 (0.872, 1.000), which demonstrated comparable performance to the machine learning model (P = .814). For Group A, 13 significant univariate variables were utilized to construct the machine learning model with an AUC of 0.917 (0.842, 0.993).Conclusion: Multiple features were associated with CSP prognosis, such as GSUCL, IMPL, RMT, and the anterior–posterior diameter of the gestational sac at the level of the niche (GSSH). The CSP prognostic prediction can be achieved by integrating machine learning and linear scoring models to balance performance and interpretability, which can assist clinicians in treatment decisions.