iNICU Intelligence leverages the computational and technological advances in augmenting the algorithms required for predictive models by integrating clinical data from EMR, real time physiological data from bedside medical devices, laboratory results and live video streaming.
Using various algorithms and machine learning models, this product aims to optimize the length of stay (LOS), predict the onset and severity of diseases in neonates prior to clinical manifestation of symptoms thereby improving the clinical outcomes.
Understanding human development leads to better conceptualization of the mechanisms associated with health and diseases. Early identification of individual disease progression can aid in the prevention of diseases and their complications. Keeping this in mind, observing the neonate’s phenotype at the time of birth, disease onset and contributory factors may help in predicting the health status of the neonate during hospital stay. This may also help in identifying the interventions to improve post discharge outcomes.
Currently used predictive models are either limited to first few hours of clinical and laboratory data or are limited to only one physiological parameter. Some of these predictive models are not even used in clinical settings, which leads to loss of important diagnostic information.
Therefore, the need of the hour is to have an integrated solution of neonatal practice with computational and technological advances for predicting the neonate’s health status. Augmenting the predictive models with clinical data, laboratory data and physiological data over time can improve the accuracy in clinical management of neonates, including the length of stay.
iNICU Intelligence is an analytics solution that is built on various algorithms leveraging the computational and technological advances required for integrating clinical data from EMR, real time physiological data from bedside medical devices and laboratory results.
Through this solution, CHIL aims to provide NICUs reliable and tested analytics to:
- Optimize the length of stay (LOS) via accurate detection of clinical events and variability in physiological parameters, deviation from nutrition and medication guidelines and bedside tablet interface providing key insights (real time & predictive).
- Early prediction of the onset, severity of diseases and mortality in neonates using big data of morbidities combined with deep learning platform to predict probability of diseases. This also involves having a feedback loop mechanism with NEO device to provide bedside alerts and notifications.
- Extracting data from wearable and live camera feeds to augment clinical decision making.
- Improved clinical care outcome by following standardized guidelines for nutrition calculations, medications and progress notes.
The solution collects, integrates and analyses digital data from EMR, laboratory and bedside medical devices over a period of time. Obtained data is segregated into different risk factor categories and this data is then utilized to optimise LOS and associated analytics, like the change in clinical state of morbidities and mortalities are displayed on the bedside tablet interface to improve the operational efficiency of clinical interventions.
The various parameters which are studied and used in the analysis of data are deviation of nutritional and medication orders from recommended guidelines, antenatal and perinatal factors, etc.
14 Robinson Road, #12-01/02 Far East Finance Building, Singapore 048545.
firstname.lastname@example.org | +91-11-40644232