Machine Learning Analysis of Readmission of Patients Diagnosed With Ischemic and Pulmonary Heart Diseases using the the data for the year 2016 from the National Readmission Database (NRD)
engineering
Description
Machine Learning Analysis of Readmission of Patients
Diagnosed With Ischemic and Pulmonary Heart Diseases using the the data for the year 2016 from the National
Readmission Database (NRD) I will need help on the literature review from peer reviewed journals on the above topic for publications from 2000 to 2019 (about 25 or more references).
Hospital readmission accounts for a statistically
significant ratio of inpatients in a hospital which increases the healthcare
cost. Various studies have referred hospital readmission as the admission
within 30 days after the initial hospital discharge, either occurring in a
different hospital or the same hospital discharged from (Yu et al. 2015; Stone
and Hoffman 2010). Additionally, it has been indicated that hospital
readmission rate is associated with patient's comorbidities, age and other
several factors such as the time take to be discharged before readmission (Wang
et al. 2014, Yu et al. 2015).
Despite hospital readmission prediction being important both
to hospital management and the entire health system, the majority of the
existing studies have poor prediction and analysis results hindering the
generalization of these methods. (Kansagara
et al. 2011). For instance, sometimes, the LACE index is used to model the risk
associated with hospital readmission in various clinical steps (Walraven et al.
2010; Gruneir et al. 2011). The Area Under receiver operation is another index
of interest in modeling hospital readmission cases. Kansagara et al. (2011),
states that the AUC is a standard index of predicting accuracy.
This research paper aims to explore the data at hand
concerning readmission cases and ease the predictability for practical use.
With enough data that is representative of the population of interest, various
machine learning models will be built which can be easily utilized by hospitals
and the general public as well as other researchers.
