Dongkai Li

Dongkai Li

Auto-generated profile, based on publicly available information

Dongkai Li

Chinese Acad Med Sci, Peking Union Med Coll Hosp, Peking Union Med Coll, Dept Crit Care Med,State Key Lab Complex Severe &, Beijing, Peoples R 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
22 June 2020 in JMIR Medical Informatics
Verified
Journal Article
Open access
LSLongxiang Su
Longxiang Su
CLChun Liu
Chun Liu
DLDongkai Li
Dongkai Li
JHJie He
Jie He
FZFanglan Zheng
Fanglan Zheng
Yun LongYun Long
Yun Long
Background: Heparin is one of the most commonly used medications in intensive care units. In clinical practice, the use of a weight-based heparin dosing nomogram is standard practice for the treatment of thrombosis. Recently, machine learning techniques have dramatically improved the ability of computers to provide clinical decision support and have allowed for the possibility of computer generated, algorithm-based heparin dosing recommendations. Objective: The objective of this study was to predict the effects of heparin treatment using machine learning methods to optimize heparin dosing in intensive care units based on the predictions. Patient state predictions were based upon activated partial thromboplastin time in 3 different ranges: subtherapeutic, normal therapeutic, and supratherapeutic, respectively. Methods: Retrospective data from 2 intensive care unit research databases (Multiparameter Intelligent Monitoring in Intensive Care III, MIMIC-III; e–Intensive Care Unit Collaborative Research Database, eICU) were used for the analysis. Candidate machine learning models (random forest, support vector machine, adaptive boosting, extreme gradient boosting, and shallow neural network) were compared in 3 patient groups to evaluate the classification performance for predicting the subtherapeutic, normal therapeutic, and supratherapeutic patient states. The model results were evaluated using precision, recall, F1 score, and accuracy. Results: Data from the MIMIC-III database (n=2789 patients) and from the eICU database (n=575 patients) were used. In 3-class classification, the shallow neural network algorithm performed the best (F1 scores of 87.26%, 85.98%, and 87.55% for data set 1, 2, and 3, respectively). The shallow neural network algorithm achieved the highest F1 scores within the patient therapeutic state groups: subtherapeutic (data set 1: 79.35%; data set 2: 83.67%; data set 3: 83.33%), normal therapeutic (data set 1: 93.15%; data set 2: 87.76%; data set 3: 84.62%), and supratherapeutic (data set 1: 88.00%; data set 2: 86.54%; data set 3: 95.45%) therapeutic ranges, respectively. Conclusions: The most appropriate model for predicting the effects of heparin treatment was found by comparing multiple machine learning models and can be used to further guide optimal heparin dosing. Using multicenter intensive care unit data, our study demonstrates the feasibility of predicting the outcomes of heparin treatment using data-driven methods, and thus, how machine learning–based models can be used to optimize and personalize heparin dosing to improve patient safety. Manual analysis and validation suggested that the model outperformed standard practice heparin treatment dosing.

20 May 2021 in Journal of Medical Internet Research
Verified
Journal Article
Open access
DLDongkai Li
Dongkai Li
JGJianwei Gao
Jianwei Gao
NHNa Hong
Na Hong
HWHao Wang
Hao Wang
LSLongxiang Su
Longxiang Su
WZWeiguo Zhu
Weiguo Zhu
Background: Unfractionated heparin is widely used in the intensive care unit as an anticoagulant. However, weight-based heparin dosing has been shown to be suboptimal and may place patients at unnecessary risk during their intensive care unit stay. Objective: In this study, we intended to develop and validate a machine learning–based model to predict heparin treatment outcomes and to provide dosage recommendations to clinicians. Methods: A shallow neural network model was adopted in a retrospective cohort of patients from the Multiparameter Intelligent Monitoring in Intensive Care III (MIMIC III) database and patients admitted to the Peking Union Medical College Hospital (PUMCH). We modeled the subtherapeutic, normal, and supratherapeutic activated partial thromboplastin time (aPTT) as the outcomes of heparin treatment and used a group of clinical features for modeling. Our model classifies patients into 3 different therapeutic states. We tested the prediction ability of our model and evaluated its performance by using accuracy, the kappa coefficient, precision, recall, and the F1 score. Furthermore, a dosage recommendation module was designed and evaluated for clinical decision support. Results: A total of 3607 patients selected from MIMIC III and 1549 patients admitted to the PUMCH who met our criteria were included in this study. The shallow neural network model showed results of F1 scores 0.887 (MIMIC III) and 0.925 (PUMCH). When compared with the actual dosage prescribed, our model recommended increasing the dosage for 72.2% (MIMIC III, 1240/1718) and 64.7% (PUMCH, 281/434) of the subtherapeutic patients and decreasing the dosage for 80.9% (MIMIC III, 504/623) and 76.7% (PUMCH, 277/361) of the supratherapeutic patients, suggesting that the recommendations can contribute to clinical improvements and that they may effectively reduce the time to optimal dosage in the clinical setting. Conclusions: The evaluation of our model for predicting heparin treatment outcomes demonstrated that the developed model is potentially applicable for reducing the misdosage of heparin and for providing appropriate decision recommendations to clinicians.

25 March 2021 in JMIR Medical Informatics
Verified
Journal Article
Open access
HJHuizhen Jiang
Huizhen Jiang
LSLongxiang Su
Longxiang Su
HWHao Wang
Hao Wang
DLDongkai Li
Dongkai Li
CZCongpu Zhao
Congpu Zhao
WZWeiguo Zhu
Weiguo Zhu
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: It is especially necessary to pay attention to the critically ill patients in ICU(Intensive Care Units) real time. Scoring systems are mostly used in the risk prediction of mortality, while usually they are not so precise and real-time with the clinical data simply weighted, and it is also time-consuming for clinical staff. Objective: We would like to fuse all the medical data together and predict the real-time mortality of ICU patients by machine learning method, which would be valuable and significant. Besides, we want to explore predicting the mortality by noninvasive data to lessen the pain of patients. Methods: In this paper, we established 5 models to predict mortality real-time based on different features. Based on monitoring data, examination data and scoring data, we structured the feature engineering. 5 Real-time Mortality prediction models were RMM(Monitoring features), RMA(APACHE and monitoring features), RMS(SOFA and monitoring features), RMME(Monitoring and Examination features) and RM(all features from monitoring, examination data and scoring data). Then, we compared the performance of all models and put more focus on the noninvasive method RMM. Results: After extensive experiments, the performance of RMME was superior to that of other 4 models. With the scoring features included, the model showed worse performance. And, RMM only based on monitoring features performed better than that of RMA and RMS. Therefore, it is meaningful and practicable to predict mortality by the noninvasive way, which could reduce the extra physical damage to patients like drawing blood. Moreover, we explored the top 9 features relevant with the real-time mortality prediction. Top 9 features were "ABP (mmHg) invasive mean pressure", "Heart rate", "ABP (mmHg) invasive systolic pressure", "Oxygen concentration", "SPO2", "Balance of inflow and outflow", "Total input", "ABP (mmHg) invasive diastolic pressure" and "NBP-average pressure", which could be paid more focus on during the general clinical work. Conclusions: This research could be helpful in real-time mortality prediction of ICU patients, especially by the noninvasive method. It is meaningful and friendly to patients, which is of strong practical significance.

19 March 2021 in BMC Anesthesiology
Verified
Journal Article
Open access
PDF file
BTBO TANG
BO TANG
LSLongxiang Su
Longxiang Su
DLDongkai Li
Dongkai Li
YWYe Wang
Ye Wang
QLQianqian Liu
Qianqian Liu
XZXiang Zhou
Xiang Zhou
To investigate the optimal target e of lactate kinetics at different time during the resuscitation, the factors that influence whether the kinetics achieve the goals, and the clinical implications of different clinical phenotypes. Patients with hyperlactatemia between May 1, 2013 and December 31, 2018 were retrospectively analyzed. Demographic data, basic organ function, hemodynamic parameters at ICU admission (T0) and at 6 h, 12 h, 24 h, 48 h, and 72 h, arterial blood lactate and blood glucose levels, cumulative clinical treatment conditions at different time points and final patient outcomes were collected. A total of 3298 patients were enrolled, and the mortality rate was 12.2%. The cutoff values of lactate kinetics for prognosis at 6 h, 12 h, 24 h, 48 h, and 72 h were 21%, 40%, 57%, 66%, and 72%. The APACHE II score, SOFA score, heart rate (HR), and blood glucose were risk factors that correlated with whether the lactate kinetics attained the target goal. Based on the pattens of the lactate kinetics, eight clinical phenotypes were proposed. The odds ratios of death for clinical phenotypes VIII, IV, and II were 4.39, 4.2, and 5.27-fold of those of clinical phenotype I, respectively. Stepwise recovery of lactate kinetics is an important resuscitation target for patients with hyperlactatemia. The APACHE II score, SOFA score, HR, and blood glucose were independent risk factors that influenced achievement of lactate kinetic targets. The cinical phenotypes of stepwise lactate kinetics are closely related to the prognosis.

12 January 2021 in JMIR Preprints
Verified
Preprint
Open access
DLDongkai Li
Dongkai Li
JGJianwei Gao
Jianwei Gao
NHNa Hong
Na Hong
HWHao Wang
Hao Wang
LSLongxiang Su
Longxiang Su
WZWeiguo Zhu
Weiguo Zhu
BACKGROUND Unfractionated heparin is widely used in the intensive care unit as an anticoagulant. However, weight-based heparin dosing has been shown to be suboptimal and may place patients at unnecessary risk during their intensive care unit stay. OBJECTIVE In this study, we intended to develop and validate a machine learning–based model to predict heparin treatment outcomes and to provide dosage recommendations to clinicians. METHODS A shallow neural network model was adopted in a retrospective cohort of patients from the Multiparameter Intelligent Monitoring in Intensive Care III (MIMIC III) database and patients admitted to the Peking Union Medical College Hospital (PUMCH). We modeled the subtherapeutic, normal, and supratherapeutic activated partial thromboplastin time (aPTT) as the outcomes of heparin treatment and used a group of clinical features for modeling. Our model classifies patients into 3 different therapeutic states. We tested the prediction ability of our model and evaluated its performance by using accuracy, the kappa coefficient, precision, recall, and the F1 score. Furthermore, a dosage recommendation module was designed and evaluated for clinical decision support. RESULTS A total of 3607 patients selected from MIMIC III and 1549 patients admitted to the PUMCH who met our criteria were included in this study. The shallow neural network model showed results of F1 scores 0.887 (MIMIC III) and 0.925 (PUMCH). When compared with the actual dosage prescribed, our model recommended increasing the dosage for 72.2% (MIMIC III, 1240/1718) and 64.7% (PUMCH, 281/434) of the subtherapeutic patients and decreasing the dosage for 80.9% (MIMIC III, 504/623) and 76.7% (PUMCH, 277/361) of the supratherapeutic patients, suggesting that the recommendations can contribute to clinical improvements and that they may effectively reduce the time to optimal dosage in the clinical setting. CONCLUSIONS The evaluation of our model for predicting heparin treatment outcomes demonstrated that the developed model is potentially applicable for reducing the misdosage of heparin and for providing appropriate decision recommendations to clinicians.

26 August 2020 in JMIR Preprints
Verified
Preprint
Open access
HJHuizhen Jiang
Huizhen Jiang
LSLongxiang Su
Longxiang Su
HWHao Wang
Hao Wang
DLDongkai Li
Dongkai Li
CZCongpu Zhao
Congpu Zhao
WZWeiguo Zhu
Weiguo Zhu
BACKGROUND Monitoring critically ill patients in intensive care units (ICUs) in real time is vitally important. Although scoring systems are most often used in risk prediction of mortality, they are usually not highly precise, and the clinical data are often simply weighted. This method is inefficient and time-consuming in the clinical setting. OBJECTIVE The objective of this study was to integrate all medical data and noninvasively predict the real-time mortality of ICU patients using a gradient boosting method. Specifically, our goal was to predict mortality using a noninvasive method to minimize the discomfort to patients. METHODS In this study, we established five models to predict mortality in real time based on different features. According to the monitoring, laboratory, and scoring data, we constructed the feature engineering. The five real-time mortality prediction models were RMM (based on monitoring features), RMA (based on monitoring features and the Acute Physiology and Chronic Health Evaluation [APACHE]), RMS (based on monitoring features and Sequential Organ Failure Assessment [SOFA]), RMML (based on monitoring and laboratory features), and RM (based on all monitoring, laboratory, and scoring features). All models were built using LightGBM and tested with XGBoost. We then compared the performance of all models, with particular focus on the noninvasive method, the RMM model. RESULTS After extensive experiments, the area under the curve of the RMM model was 0.8264, which was superior to that of the RMA and RMS models. Therefore, predicting mortality using the noninvasive method was both efficient and practical, as it eliminated the need for extra physical interventions on patients, such as the drawing of blood. In addition, we explored the top nine features relevant to real-time mortality prediction: invasive mean blood pressure, heart rate, invasive systolic blood pressure, oxygen concentration, oxygen saturation, balance of input and output, total input, invasive diastolic blood pressure, and noninvasive mean blood pressure. These nine features should be given more focus in routine clinical practice. CONCLUSIONS The results of this study may be helpful in real-time mortality prediction in patients in the ICU, especially the noninvasive method. It is efficient and favorable to patients, which offers a strong practical significance.