Classifying inpatient perception of hospitalization experience across the US 50 states using the convolutional neural networks [post]

2020 unpublished
Backgrounds The grade of hospital-service quality is required for the classi cation which is never applicable in the literature. We aimed to classify the grade of inpatients' perceptions of patients' hospitalization experience of states in the US using convolutional neural networks(CNN).Methods We downloaded HCAHPS (Hospital Consumer Assessment of Healthcare Providers and Services) data from the 2007-2014 summaries of survey results. The data collection was carried out to the k-mean and the CNN
more » ... k-mean and the CNN that were used as unsupervised and supervised learnings for (1) dividing the US sates into two classes (n = 19 and 32 of lower and higher grades) and (2) building a hospital service predictive model to estimate 38 parameters. We calculated the sensitivity, speci city, and receiver operating characteristic curve [area under the curve (AUC)] across studies for comparison. An app predicting the hospital service for a speci c US state was developed involving the model's 38 estimated parameters for a website assessment. Results We observed that (1) the two-year 20-item model yields a higher accuracy rate (0.98) with an AUC (0.99; 95% CI 0.95-1.00) based on the 51 states; and (2) an available app for predicting hospital-service quality was successfully developed and demonstrated in this study. A smartphone app was designed to classify the grade of hospital service for each US state. Conclusions The 20-item model with the 38 parameters estimated by using CNN for improving the accuracy of the grade about the inpatients' perceptions of patients' hospitalization experience of states in the US for hospital service. An app developed for helping patients' self-assess hospital service quality in each US state is required for application in the future. Highlights We developed an App in use for assessing the quality of hospital service in the US states. The CNN algorithm was used in this study, particularly in MS Excel, which is rarely seen in the literature. A dashboard on Google maps successfully reports inpatient perception of hospitalization experience across US 50 states. A smartphone APP was designed to get feedback directly from patients' perceptions of hospitalization experience.
doi:10.21203/rs.2.20242/v1 fatcat:rfcm72kmfjctjjmlssqnxypjnq