Related Experiment Video
Updated: Oct 11, 2025
Point-Of-Care Ultrasound Screening for Proximal Lower Extremity Deep Venous Thrombosis
Published on: February 10, 2023
Prediction of central venous catheter-associated deep venous thrombosis in pediatric critical care settings
Haomin Li1, Yang Lu2, Xian Zeng2
1Clinical Data Center, The Children's Hospital, Zhejiang University School of Medicine and National Clinical Research Center for Child Health, 3333 Binsheng Road, Hangzhou, 310052, China. hmli@zju.edu.cn.
Insights
A new AI model accurately predicts central venous catheter-associated deep venous thrombosis (CADVT) in children up to 72 hours in advance. This advance prediction of CADVT in pediatric patients can improve thromboprophylaxis measures.
Area of Science:
- Medical Informatics
- Artificial Intelligence in Medicine
- Pediatric Thrombosis
Background:
- Central venous catheter (CVC)-associated deep venous thrombosis (CADVT) incidence is rising in pediatric patients.
- Current screening methods for venous thromboembolism risk in hospitalized children have low sensitivity for CADVT.
- This study addresses the need for improved CADVT prediction in pediatric intensive care units.
Purpose of the Study:
- To develop and evaluate a multimodal deep learning (MMDL) model for predicting CADVT in pediatric patients.
- To compare the MMDL model's performance against traditional machine learning models and existing MMDL approaches.
- To assess the model's ability to predict CADVT at various time points before clinical detection.
Main Methods:
- A multimodal deep learning (MMDL) model utilizing LSTM and GRUs was developed.
- Data from 1830 pediatric ICU patients with CVC placement were analyzed, including demographics, clinical conditions, labs, vitals, and medications.
- Model performance was evaluated using accuracy, recall, AUC, and AP at 24, 48, and 72 hours pre-CADVT, with comparisons to LR, RF, GBDT, and a benchmark MMDL.
Main Results:
- The proposed MMDL model significantly outperformed benchmark models in predicting CADVT at all evaluated time points.
- The model achieved 77% accuracy and 90% recall at 24 hours prior to CADVT detection.
- The model demonstrated robust predictive capability up to 72 hours in advance, with >75% accuracy, >87% recall, and an AUC of 0.82.
Conclusions:
- A machine learning-based method was successfully developed for the early prediction of CADVT in pediatric patients.
- The findings suggest that artificial intelligence (AI) can offer effective thromboprophylaxis strategies in pediatric intensive care.
- The developed AI model shows potential for proactive management of CADVT in high-risk pediatric populations.
Background:
An increase in the incidence of central venous catheter (CVC)-associated deep venous thrombosis (CADVT) has been reported in pediatric patients over the past decade. At the same time, current screening guidelines for venous thromboembolism risk have low sensitivity for CADVT in hospitalized children. This study utilized a multimodal deep learning model to predict CADVT before it occurs.
Methods:
Children who were admitted to intensive care units (ICUs) between December 2015 and December 2018 and with CVC placement at least 3 days were included. The variables analyzed included demographic characteristics, clinical conditions, laboratory test results, vital signs and medications. A multimodal deep learning (MMDL) model that can handle temporal data using long short-term memory (LSTM) and gated recurrent units (GRUs) was proposed for this prediction task. Four benchmark machine learning models, logistic regression (LR), random forest (RF), gradient boosting decision tree (GBDT) and a published cutting edge MMDL, were used to compare and evaluate the models with a fivefold cross-validation approach. Accuracy, recall, area under the ROC curve (AUC), and average precision (AP) were used to evaluate the discrimination of each model at three time points (24 h, 48 h and 72 h) before CADVT occurred. Brier score and Spiegelhalter's z test were used measure the calibration of these prediction models.
Results:
A total of 1830 patients were included in this study, and approximately 15% developed CADVT. In the CADVT prediction task, the model proposed in this paper significantly outperforms both traditional machine learning models and existing multimodal deep learning models at all 3 time points. It achieved 77% accuracy and 90% recall at 24 h before CADVT was discovered. It can be used to accurately predict the occurrence of CADVT 72 h in advance with an accuracy of greater than 75%, a recall of more than 87%, and an AUC value of 0.82.
Conclusion:
In this study, a machine learning method was successfully established to predict CADVT in advance. These findings demonstrate that artificial intelligence (AI) could provide measures for thromboprophylaxis in a pediatric intensive care setting.
Related Concept Videos
01:29Venous Thrombosis III: Interprofessional Care
01:20Venous Thrombosis II: Clinical Manifestations and Diagnostic Studies
01:30Venous Thrombosis I: Introduction
01:27Parentral Nutrition: Centeral and Peripheral Parental Nutrition
PN can be administered through two primary routes:
1. Central Parenteral Nutrition (CPN):
CPN involves delivering a high concentration of nutrients through a large vein. This is typically achieved using a Peripherally Inserted Central Catheter (PICC) or,...
01:30Venous Thrombosis IV: Nursing Management
01:29Pulmonary Embolism II: Diagnostic Studies and Interprofessional Care