I-Corps: Data Completeness and Inconsistency Analysis Platform

点击翻译按钮获取中文摘要

This I-Corps project will impact a range of healthcare industry jobs, increasing financial efficiency and reduce data-driven misdiagnosis and mistreatment. Healthcare patient outcomes and experiences will improve with reductions in medication errors, misdiagnosis, and other inaccuracies that can cause harm to patients and create liability issues with healthcare providers. Processes associated with acquiring electronic health data are often function slowly, incompletely, and often without the full consent of the patient or adequate metadata to track critical information required for treating a patient. The solution developed here can potentially enable proper data correctness and address inconsistencies which can lead to better patient outcomes and improve legal liability framework.This I-Corps project provides intellectual merits that will improve the healthcare information industry. This includes synthesis of mathematical models and algorithms that analyze data completeness and consistency. Specifically, this will lead to a hybrid machine learning approach that will utilize unsupervised learning to automatically classify various electronic health records as complete or incomplete, and supervised learning to confirm such classification and further grade the likelihood that the entered data is accurate and correct. Furthermore, this will also lead to evolutions in applications of reinforcement learning supporting greater accuracy, and anomaly detection. Overall, this project will lead to improving the quality of healthcare delivery and innovations in supporting evolutions in data science.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.