Yu Zhang

Yu Zhang

Auto-generated profile, based on publicly available information

Yu Zhang

Institute of Medical Information and Library, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, China

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
17 December 2018 in JMIR Medical Informatics
Verified
Journal Article
Open access
YZYu Zhang
Yu Zhang
XWXuwen Wang
Xuwen Wang
ZHZhen Hou
Zhen Hou
JLJiao Li
Jiao Li
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Electronic health records (EHR) is an important data resource for clinical studies and applications. Physicians or clinicians describe patients’ disorders or treatment procedures using free texts in EHR. The narrative information play an important role in patient treatment and clinical research. However, it is challenging to made machine understand the clinical narratives. Objective: This study aimed to automatically identify Chinese clinical entities from free texts in EHR, and make machine semantically understand diagnosis, test, body part, symptom, treatment and, etc. Methods: The data set we used for this study is the benchmark data set with human annotated Chinese electronic health records, released by CCKS 2017 CNER challenge task. Two machine learning (ML) models, conditional random fields (CRF) method and bidirectional LSTM-CRF, were applied to recognize clinical entities from Chinese EHR data. For training the CRF-based model, we selected features as bag of Chinese characters, part-of-speech (POS) tags, character types and the position of characters. For the bidirectional LSTM-CRF-based model, character embeddings and segmentation information were used as features. In addition, we also employed a dictionary-based approach as the baseline for performance evaluation purpose. Precision, Recall and the harmonic average of the precision and recall (F1 score) were used to evaluate the performance of the methods. Results: Experiments on the test set showed that our methods were able to automatically identify types of Chinese clinical entities such as diagnosis, test, symptom body part and treatment in one-round running. The identification overall performance of CRF and bidirectional LSTM-CRF achieved Precision of 0.9203 and 0.9112, Recall of 0.8709 and 0.8974, F1 score of 0.8949 and 0.9043 respectively. The result also indicated that our methods performed well on recognizing each type of clinical entities, in which the “symptom” type achieved the best with F1 score over 0.96. Moreover, as the number of features increased, F1 score of the CRF model increases from of 0.8547 to 0.8949. Conclusions: In this study, we employed two computational methods to simultaneously identify types of Chinese clinical entities from free texts in EHRs. Via training, it can effectively identify various types of clinical entities (e.g., symptom and treatment) with high accuracy. The deep learning model, bidirectional LSTM-CRF, can achieve better performance than the CRF model with little feature engineering. This study contributed to translating human-readable health information into machine-readable one.

30 November 2018 in Journal of Medical Internet Research
Verified
Journal Article
Open access
XNXu Na
Xu Na
HGHaihong Guo
Haihong Guo
YZYu Zhang
Yu Zhang
LSLiu Shen
Liu Shen
SWSizhu Wu
Sizhu Wu
JLJiao Li
Jiao Li
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet

30 January 2018 in JMIR Preprints
Verified
Preprint
Open access
YZYu Zhang
Yu Zhang
XWXuwen Wang
Xuwen Wang
ZHZhen Hou
Zhen Hou
JLJiao Li
Jiao Li
BACKGROUND Electronic health records (EHRs) are important data resources for clinical studies and applications. Physicians or clinicians describe patients’ disorders or treatment procedures in EHRs using free text (unstructured) clinical notes. The narrative information plays an important role in patient treatment and clinical research. However, it is challenging to make machines understand the clinical narratives. OBJECTIVE This study aimed to automatically identify Chinese clinical entities from free text in EHRs and make machines semantically understand diagnoses, tests, body parts, symptoms, treatments, and so on. METHODS The dataset we used for this study is the benchmark dataset with human annotated Chinese EHRs, released by the China Conference on Knowledge Graph and Semantic Computing 2017 clinical named entity recognition challenge task. Overall, 2 machine learning models, the conditional random fields (CRF) method and bidirectional long short-term memory (LSTM)-CRF, were applied to recognize clinical entities from Chinese EHR data. To train the CRF–based model, we selected features such as bag of Chinese characters, part-of-speech tags, character types, and the position of characters. For the bidirectional LSTM-CRF–based model, character embeddings and segmentation information were used as features. In addition, we also employed a dictionary-based approach as the baseline for the purpose of performance evaluation. Precision, recall, and the harmonic average of precision and recall (F1 score) were used to evaluate the performance of the methods. RESULTS Experiments on the test set showed that our methods were able to automatically identify types of Chinese clinical entities such as diagnosis, test, symptom, body part, and treatment simultaneously. With regard to overall performance, CRF and bidirectional LSTM-CRF achieved a precision of 0.9203 and 0.9112, recall of 0.8709 and 0.8974, and F1 score of 0.8949 and 0.9043, respectively. The results also indicated that our methods performed well in recognizing each type of clinical entity, in which the “symptom” type achieved the best F1 score of over 0.96. Moreover, as the number of features increased, the F1 score of the CRF model increased from 0.8547 to 0.8949. CONCLUSIONS In this study, we employed two computational methods to simultaneously identify types of Chinese clinical entities from free text in EHRs. With training, these methods can effectively identify various types of clinical entities (eg, symptom and treatment) with high accuracy. The deep learning model, bidirectional LSTM-CRF, can achieve better performance than the CRF model with little feature engineering. This study contributed to translating human-readable health information into machine-readable information.

22 July 2018 in JMIR Preprints
Verified
Preprint
Open access
XNXu Na
Xu Na
HGHaihong Guo
Haihong Guo
YZYu Zhang
Yu Zhang
LSLiu Shen
Liu Shen
SWSizhu Wu
Sizhu Wu
JLJiao Li
Jiao Li
BACKGROUND The financial relationship between physicians and industries has become a hotly debated issue globally. The Physician Payments Sunshine Act of the US Affordable Care Act (2010) promoted transparency of the transactions between industries and physicians by making remuneration data publicly accessible in the Open Payments Program database. Meanwhile, according to the World Health Organization, the majority of all noncommunicable disease deaths were caused by cardiovascular disease. OBJECTIVE This study aimed to investigate the distribution of non-research and non-ownership payments made to thoracic surgeons, to explore the regularity of financial relationships between industries and thoracic surgeons. METHODS Annual statistical data were obtained from the Open Payments Program general payment dataset from 2014-2016. We characterized the distribution of annual payments with single payment transactions greater than US $10,000, quantified the major expense categories (eg, Compensation, Consulting Fees, Travel and Lodging), and identified the 30 highest-paying industries. Moreover, we drew out the financial relations between industries to thoracic surgeons using chord diagram visualization. RESULTS The three highest categories with single payments greater than US $10,000 were Royalty or License, Compensation, and Consulting Fees. Payments related to Royalty or License transferred from only 5.38% of industries to 0.75% of surgeons with the highest median (US $13,753, $11,992, and $10,614 respectively) in 3-year period. In contrast, payments related to Food and Beverage transferred from 93.50% of industries to 98.48% of surgeons with the lowest median (US $28, $27, and $27). The top 30 highest-paying industries made up approximately 90% of the total payments (US $21,036,972, $23,304,996, and $28,116,336). Furthermore, just under 9% of surgeons received approximately 80% of the total payments in each of the 3 years. Specifically, the 100 highest cumulative payments, accounting for 52.69% of the total, transferred from 27 (6.05%) pharmaceutical industries to 86 (1.89%) thoracic surgeons from 2014-2016; 7 surgeons received payments greater than US $1,000,000; 12 surgeons received payments greater than US $400,000. The majority (90%) of these surgeons received tremendous value from only one industry. CONCLUSIONS There exists a great discrepancy in the distribution of payments by categories. Royalty or License Fees, Compensation, and Consulting Fees are the primary transferring channels of single large payments. The massive transfer from industries to surgeons has a strong “apical dominance” and excludability. Further research should focus on discovering the fundamental driving factors for the strong concentration of certain medical devices and how these payments will affect the industry itself.