Cheng Gao1, Sarah Osmundson2, Digna R Velez Edwards3, Gretchen Purcell Jackson4, Bradley A Malin5, You Chen6. 1. Department of Biomedical Informatics, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA. 2. Department of Obstetrics and Gynecology, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA. 3. Department of Biomedical Informatics, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Obstetrics and Gynecology, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA. 4. Department of Biomedical Informatics, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA; Departments of Pediatric Surgery and Pediatrics, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA; Evaluation Research Center, IBM Watson Health, Cambridge, MA, USA. 5. Department of Biomedical Informatics, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Biostatistics, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Electrical Engineering & Computer Science, School of Engineering, Vanderbilt University, Nashville, TN, USA. 6. Department of Biomedical Informatics, School of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA. Electronic address: you.chen@vanderbilt.edu.
Abstract
OBJECTIVE: Models for predicting preterm birth generally have focused on very preterm (28-32 weeks) and moderate to late preterm (32-37 weeks) settings. However, extreme preterm birth (EPB), before the 28th week of gestational age, accounts for the majority of newborn deaths. We investigated the extent to which deep learning models that consider temporal relations documented in electronic health records (EHRs) can predict EPB. STUDY DESIGN: EHR data were subject to word embedding and a temporal deep learning model, in the form of recurrent neural networks (RNNs) to predict EPB. Due to the low prevalence of EPB, the models were trained on datasets where controls were undersampled to balance the case-control ratio. We then applied an ensemble approach to group the trained models to predict EPB in an evaluation setting with a nature EPB ratio. We evaluated the RNN ensemble models with 10 years of EHR data from 25,689 deliveries at Vanderbilt University Medical Center. We compared their performance with traditional machine learning models (logistical regression, support vector machine, gradient boosting) trained on the datasets with balanced and natural EPB ratio. Risk factors associated with EPB were identified using an adjusted odds ratio. RESULTS: The RNN ensemble models trained on artificially balanced data achieved a higher AUC (0.827 vs. 0.744) and sensitivity (0.965 vs. 0.682) than those RNN models trained on the datasets with naturally imbalanced EPB ratio. In addition, the AUC (0.827) and sensitivity (0.965) of the RNN ensemble models were better than the AUC (0.777) and sensitivity (0.819) of the best baseline models trained on balanced data. Also, risk factors, including twin pregnancy, short cervical length, hypertensive disorder, systemic lupus erythematosus, and hydroxychloroquine sulfate, were found to be associated with EPB at a significant level. CONCLUSION: Temporal deep learning can predict EPB up to 8 weeks earlier than its occurrence. Accurate prediction of EPB may allow healthcare organizations to allocate resources effectively and ensure patients receive appropriate care.
OBJECTIVE: Models for predicting preterm birth generally have focused on very preterm (28-32 weeks) and moderate to late preterm (32-37 weeks) settings. However, extreme preterm birth (EPB), before the 28th week of gestational age, accounts for the majority of newborn deaths. We investigated the extent to which deep learning models that consider temporal relations documented in electronic health records (EHRs) can predict EPB. STUDY DESIGN: EHR data were subject to word embedding and a temporal deep learning model, in the form of recurrent neural networks (RNNs) to predict EPB. Due to the low prevalence of EPB, the models were trained on datasets where controls were undersampled to balance the case-control ratio. We then applied an ensemble approach to group the trained models to predict EPB in an evaluation setting with a nature EPB ratio. We evaluated the RNN ensemble models with 10 years of EHR data from 25,689 deliveries at Vanderbilt University Medical Center. We compared their performance with traditional machine learning models (logistical regression, support vector machine, gradient boosting) trained on the datasets with balanced and natural EPB ratio. Risk factors associated with EPB were identified using an adjusted odds ratio. RESULTS: The RNN ensemble models trained on artificially balanced data achieved a higher AUC (0.827 vs. 0.744) and sensitivity (0.965 vs. 0.682) than those RNN models trained on the datasets with naturally imbalanced EPB ratio. In addition, the AUC (0.827) and sensitivity (0.965) of the RNN ensemble models were better than the AUC (0.777) and sensitivity (0.819) of the best baseline models trained on balanced data. Also, risk factors, including twin pregnancy, short cervical length, hypertensive disorder, systemic lupus erythematosus, and hydroxychloroquine sulfate, were found to be associated with EPB at a significant level. CONCLUSION: Temporal deep learning can predict EPB up to 8 weeks earlier than its occurrence. Accurate prediction of EPB may allow healthcare organizations to allocate resources effectively and ensure patients receive appropriate care.
Authors: Ilia Rattsev; Natalie Flaks-Manov; Angie C Jelin; Jiawei Bai; Casey Overby Taylor Journal: J Am Med Inform Assoc Date: 2022-01-12 Impact factor: 7.942
Authors: Camilo Espinosa; Martin Becker; Ivana Marić; Ronald J Wong; Gary M Shaw; Brice Gaudilliere; Nima Aghaeepour; David K Stevenson Journal: Trends Mol Med Date: 2021-02-08 Impact factor: 15.272