No abstract

Objectives The objectives of this study are to construct the high definition phenotype (HDP), a novel time-series data structure composed of both primary and derived parameters, using heterogeneous clinical sources and to determine whether different predictive models can utilize the HDP in the neonatal intensive care unit (NICU) to improve neonatal mortality prediction in clinical settings. Materials and Methods A total of 49 primary data parameters were collected from July 2018 to May 2020 from eight level-III NICUs. From a total of 1546 patients, 757 patients were found to contain sufficient fixed, intermittent, and continuous data to create HDPs. Two different predictive models utilizing the HDP, one a logistic regression model (LRM) and the other a deep learning long–short-term memory (LSTM) model, were constructed to predict neonatal mortality at multiple time points during the patient hospitalization. The results were compared with previous illness severity scores, including SNAPPE, SNAPPE-II, CRIB, and CRIB-II. Results A HDP matrix, including 12 221 536 minutes of patient stay in NICU, was constructed. The LRM model and the LSTM model performed better than existing neonatal illness severity scores in predicting mortality using the area under the receiver operating characteristic curve (AUC) metric. An ablation study showed that utilizing continuous parameters alone results in an AUC score of >80% for both LRM and LSTM, but combining fixed, intermittent, and continuous parameters in the HDP results in scores >85%. The probability of mortality predictive score has recall and precision of 0.88 and 0.77 for the LRM and 0.97 and 0.85 for the LSTM. Conclusions and Relevance The HDP data structure supports multiple analytic techniques, including the statistical LRM approach and the machine learning LSTM approach used in this study. LRM and LSTM predictive models of neonatal mortality utilizing the HDP performed better than existing neonatal illness severity scores. Further research is necessary to create HDP–based clinical decision tools to detect the early onset of neonatal morbidities.

Neonatal period represents first 28 days of life, which is the most vulnerable time for a child’s survival especially for the preterm babies. High neonatal mortality is a prominent and persistent problem across the globe. Non-availability of trained staff and infrastructure are the major recognized hurdles in the quality care of these neonates. Hourly progress growth charts and reports are still maintained manually by nurses along with continuous calculation of drug dosage and nutrition as per the changing weight of the baby. iNICU (integrated Neonatology Intensive Care Unit) leverages Beaglebone and Intel Edison based IoT integration with biomedical devices in NICU i.e. monitor, ventilator and blood gas machine. iNICU is hosted on IBM Softlayer based cloud computing infrastructure and map NICU workflow in Java based responsive web application to provide translational research informatics support to the clinicians. iNICU captures real time vital parameters i.e. respiration rate, heart rate, lab data and PACS amounting for millions of data points per day per child. Stream of data is sent to Apache Kafka layer which stores the same in Apache Cassandra NoSQL. iNICU also captures clinical data like feed intake, urine output, and daily assessment of child in PostgreSQL database. It acts as first Big Data hub (of both structured and unstructured data) of neonates across India offering temporal (longitudinal) data of their stay in NICU and allow clinicians in evaluating efficacy of their interventions. iNICU leverages drools based clinical rule based engine and deep learning based big data analytical model coded in R and PMML. iNICU solution aims to improve care time, fills skill gap, enable remote monitoring of neonates in rural regions, assists in identifying the early onset of disease, and reduction in neonatal mortality.Electronic supplementary materialThe online version of this article (doi:10.1007/s10916-017-0774-8) contains supplementary material, which is available to authorized users.

Increased length of stay (LOS) in intensive care units is directly associated with the financial burden, anxiety, and increased mortality risks. In the current study, we have incorporated the association of day-to-day nutrition and medication data of the patient during its stay in hospital with its predicted LOS. To demonstrate the same, we developed a model to predict the LOS using risk factors (a) perinatal and antenatal details, (b) deviation of nutrition and medication dosage from guidelines, and (c) clinical diagnoses encountered during NICU stay. Data of 836 patient records (12 months) from two NICU sites were used and validated on 211 patient records (4 months). A bedside user interface integrated with EMR has been designed to display the model performance results on the validation dataset. The study shows that each gestation age group of patients has unique and independent risk factors associated with the LOS. The gestation is a significant risk factor for neonates < 34 weeks, nutrition deviation for < 32 weeks, and clinical diagnosis (sepsis) for ≥ 32 weeks. Patients on medications had considerable extra LOS for ≥ 32 weeks’ gestation. The presented LOS model is tailored for each patient, and deviations from the recommended nutrition and medication guidelines were significantly associated with the predicted LOS.

Background Critical care units (CCUs) with extensive use of various monitoring devices generate massive data. To utilize the valuable information of these devices; data are collected and stored using systems like clinical information system and laboratory information management system. These systems are proprietary, allow limited access to their database and, have the vendor-specific clinical implementation. In this study, we focus on developing an open-source web-based meta-data repository for CCU representing stay of the patient with relevant details. Methods After developing the web-based open-source repository named data dictionary (DD), we analyzed prospective data from 2 sites for 4 months for data quality dimensions (completeness, timeliness, validity, accuracy, and consistency), morbidity, and clinical outcomes. We used a regression model to highlight the significance of practice variations linked with various quality indicators. Results DD with 1555 fields (89.6% categorical and 11.4% text fields) is presented to cover the clinical workflow of a CCU. The overall quality of 1795 patient days data with respect to standard quality dimensions is 87%. The data exhibit 88% completeness, 97% accuracy, 91% timeliness, and 94% validity in terms of representing CCU processes. The data scores only 67% in terms of consistency. Furthermore, quality indicators and practice variations are strongly correlated (P &lt; 0.05). Conclusion This study documents DD for standardized data collection in CCU. DD provides robust data and insights for audit purposes and pathways for CCU to target practice improvements leading to specific quality improvements.