��������

NSF Grant Will Support Health Care Delivery Network Systems

---

Researchers from ��������’s have received a $599,983 grant from the National Science Foundation to conduct research on data mining models and algorithms for complex dynamic information networks for health care delivery.

Health information systems encompass patient health records, patient preferences and values, the status of clinical orders, administrative information, and a broad range of process and systems performance data needed to diagnose and prescribe treatment. These systems also are used to analyze, control and optimize how these delivery systems and subsystems perform.   

“Patient-centered care is key to high quality care. Information and information exchange of the health care needs of an individual are crucial for all levels of the health care delivery system from the patient to the care team to the health care organization,” said , Ph.D., dean of ��������’s College of Engineering and Computer Science. “This National Science Foundation grant will help our research team to enhance how care providers capture, tap into, and integrate vital information streams for patient-centered care.”

The project, titled “KMELIN: Knowledge Mining and Embedding Learning for Complex Dynamic Information Networks,” will enable network analytics, modeling, and decision-making processes to take full advantage of network relationships. Because object contents are relatively static and stable, whereas relationships can evolve quickly and change dynamically, this research provides a game changing paradigm shift from content-oriented to relationship-oriented. This platform is more suitable for customized treatment planning, individual care, consolidated care coordination, and decision making.

“Although data relationships have become more rich and comprehensive than ever before, existing systems are mostly relational-database driven and hard to integrate complex relationships of networked data for big data analytics,” said , Ph.D., principal investigator and a professor, who is collaborating with co-principal investigators , Ph.D., a professor, and , Ph.D., an assistant professor, all in the . “By designing new algorithms and framework for knowledge sharing and decision support, we will help to shift existing health information systems from traditional databases toward becoming network centered systems.”

Complex dynamic information networks (CDIN)consist of data objects that are highly correlated with a variety of dependency relationships, such as patient-physician interactions or patient-medication-insurance claims. Each data object as a CDIN node has rich contents, such as biometric information of a patient, disease symptoms, or hospital logistics. Data objects and their relationships also continuously evolve and change.  

“Many health, social, physical, and biological systems share complex dynamic information networks,” said Zhu. “These networks involve rich node content information as well as complex and dynamic linkages between nodes. Finding patterns or knowledge from these networks has been a significant challenge.”

Zhu, Agarwal and Wang will use automatic network profiling, as well as network node classification approaches for automated hospital re-admission diagnoses. These approaches will help them to form a set of tools that will ensure that knowledge can be rapidly discovered from dynamic changing networks.

“With electronic health records in wide use today, the research that professors Zhu, Agarwal and Wang are conducting will chart a new direction and path for the exchange of information within complex health care systems that will profoundly improve patient care and processes,” said , Ph.D., chair and professor in ��������’s Department of Computer and Electrical Engineering and Computer Science.

The project’s objectives also include training three Ph.D. students, publishing high quality publications in the field of health decision informatics, data mining, and big data analytics, and developing a new course based on information networks and social networks, and data mining models for big data analytics.

-��������-

Tags: faculty and staff | research | science | technology | engineering