Clinical Decision Support System for maintaining Distributed Electronic Health Records. Test phase 1, test phase 2, srs, design phase and coding final deliverable

Project Domain / Category

Data warehouse Application by using data mining and Artificial Intelligence (AI) technique.

Abstract/Introduction

Building Clinical Decision Support System (CDSS) will improve the quality and efficiency of healthcare. Improving the quality of healthcare, reducing medical errors and guarantying the safety of patients are the most serious duty of the hospital. Electronic Health Record (EHR) is used to achieve these goals. Electronic Health Record (HER) includes XRays, ECG,  MRI reports  etc.  An electronic health record (EHR), is the systematized collection of patient and population electronically-stored in a digital format. These records can be shared across different health care sites. EHR has a very large data source which can guide and improve the clinical decision making process. Clinical Decision Support System (CDSS) satisfies the compatibility, interoperability, and scalability. The CDSS will take advantages of Electronic Health Record (EHR), data mining techniques, clinical databases, domain experts’ knowledge bases, and available technologies and standards to provide decision making support for the healthcare personnel. Clinical Decision Support System (CDSS) contains a set of knowledge bases (one in each hospital). It can be extracted offline from domain experts. CDSS only depends on these knowledge bases. It can be inactive and will become not applicable. The solution is to continually update these knowledge bases to make CDSS more active. At each site, new knowledge will be discovered and added to knowledge base and new expert knowledge will be discovered. Data mining engine will be connected to local Electronic Health Record EHR and clinical databases. This action will make CDSS more active by including the most recent knowledge from active databases.

Functional Requirements:

The distributed system will be co-operative and is integrated knowledge bases. Each knowledge base in each hospital will be in specific domain. At each hospital, CDSS will build patient profile from patient’s medical history and current diagnose, and it will use its local knowledge base to make decision. The goal of this distributed CDSS is to perform following activities.

• In an on-line operation, Healthcare personnel enters patient Universal ID (UID) which identify the patient nation-widely, and enters subject data or current diagnose (i.e., healthcare data that needs decision making).
• If the patient has record then the service returns message indicating the patient record. It indicates the last N visits, visits within specific period and specific disease’s related data, etc.
• The returned records will be collected and filtered to produce the patient
• CDSS which makes inference of diagnose and determine the correct
• CDSS can be programmed by any AI methodology as artificial neural It can access, query, and interpret the data and knowledge.
• This way CDSS will take decision based on the initial physician diagnose, EHRs, and knowledge from its local Knowledge bases (KB) and other Knowledge bases KBs. Knowledge base (KB) contains the most recent knowledge. Each KB will be specialized in specific

 

• Build co-operative knowledge bases from different domain experts’ knowledge and most

recent academic researches.

• The final results will be displayed to healthcare personnel or physician according to the level of automation in
• You can standardize knowledge into XML format before
• Connect CDSS to EHR (Electronic Health Record) and clinical databases to continuously mine the most recent and applicable knowledge and adds it to local knowledge
• CDSS can consult specialized knowledge bases in other institutions for other relevant knowledge.
• Before starting to take decision, CDSS collects all patient EHRs from all sites, integrates it with current diagnose and then standardizes

 

Tools: Oracle BI Enterprise Edition, SQL server 2008, IBM InfoSphere.

 

Supervisor: Name: Muhammad Umar Farooq Email ID: umarfarooq@vu.edu.pk

Skype ID: live:umarvc