Scribd
Upload
54 views

Using Data Mining To Predict Hospital Admissions From The Emergency Department

This document summarizes a student's MCA IV semester project on using data mining techniques to predict hospital admissions from the emergency department. The student will use machine learning algorithms like logistic regression, decision trees, and gradient boosted machines to build predictive models from administrative health data. The proposed system aims to reduce emergency department crowding by identifying patients at high risk of admission. This could help improve patient flow and resource planning. Hardware requirements include a laptop or desktop computer and software requirements include Windows 10, Apache Tomcat, MySQL, and languages like HTML, CSS, JSP.

Uploaded by

Abhi S
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
54 views

Using Data Mining To Predict Hospital Admissions From The Emergency Department

This document summarizes a student's MCA IV semester project on using data mining techniques to predict hospital admissions from the emergency department. The student will use machine learning algorithms like logistic regression, decision trees, and gradient boosted machines to build predictive models from administrative health data. The proposed system aims to reduce emergency department crowding by identifying patients at high risk of admission. This could help improve patient flow and resource planning. Hardware requirements include a laptop or desktop computer and software requirements include Windows 10, Apache Tomcat, MySQL, and languages like HTML, CSS, JSP.

Uploaded by

Abhi S
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

MCA IV Semester Project I Review Presentation

Student Name :
 

Roll No
 
:

Seminar Title : Using Data Mining to Predict Hospital Admissions


 

from the Emergency Department

Guide Name
 
:
Contents
 ABSTRACT

 INTRODUCTION

 PROBLEM DEFINITION

 EXISTING SYSTEM

DISADVANTAGES OF EXISTING SYSTEM

 PROPOSED SYSTEM

ADVANTAGES OF PROPOSED SYSTEM

 SOFTWARE AND HARDWARE REQUIREMENTS


ABSTRACT

Crowding within emergency departments (EDs) can have significant


negative consequences for patients. EDs therefore need to explore the use of
innovative methods to improve patient flow and prevent overcrowding. One
potential method is the use of data mining using machine learning techniques
to predict ED admissions. This project uses routinely collected
administrative data (more records) from two major acute hospitals in
Northern Ireland to compare contrasting machine learning algorithms in
predicting the risk of admission from the ED and use three algorithms to
build the predictive models: 1) logistic regression; 2) decision trees; and 3)
gradient boosted machines (GBM).
INTRODUCTION
Emergency department (ED) crowding can have serious negative
consequences for patients and staff, such as increased wait time, ambulance
diversion, reduced staff morale, adverse patient outcomes such as increased
mortality, and cancellation of elective procedures.

Previous research has shown ED crowding to be a significant international


problem making it crucial that innovative steps are taken to address the
problem.

There are a range of possible causes of ED crowding depending on the


context, with some of the main reasons including increased ED attendances,
inappropriate attendances, a lack of alternative treatment options, a lack of
inpatient beds, ED staffing shortages, and closure of other local ED
departments .
PROBLEM DEFINITION
 
 

This Project
 
highlights the potential utility of three common machine learning
algorithms in predicting patient admissions. Practical implementation of the models
developed in this project in decision support tools would provide a snapshot of
predicted admissions from the ED at a given time, allowing for advance resource
planning and the avoidance bottlenecks in patient flow
EXISTING SYSTEM

Sun developed a logistic regression model using two years of routinely


collected administrative data to predict the probability of admission at the
point of triage.

Risk of admission was related to age, ethnicity, arrival mode, patient


acuity score, existing chronic conditions, and prior ED attendances or
admission in the past three months.

Although their data showed the admission of more females than males,
sex was not significant in the final model.

Similarly, Cameron developed a logistic regression model to predict the


probability of admissions at triage, using two years of routine
administration data collected from hospitals in Glasgow.
DISADVANTAGES OF EXISTING SYSTEM

There are no Methods for Predicting admissions based on the test data
set.

There is no Data Set extraction using fast techniques


PROPOSED SYSTEM

The proposed system implemented to reduce ED crowding and improve


patient how is the use of data mining to identify patients at high risk of an
inpatient admission, therefore allowing measures to be taken to avoid
bottlenecks in the system.

Cameron also propose that the implementation of the system could help to
improve patient satisfaction by providing the patient with advance notice that
admission is likely.

 Such a model could be developed using data mining techniques, which


involves examining and analyzing data to extract useful information and
knowledge on which decisions can be taken.

This study focuses on the use of machine learning algorithms to develop


models to predict hospital admissions from the emergency department
ADVANTAGES OF PROPOSED SYSTEM

Huge Data Set extraction using fast techniques.

Data cleansing and feature engineering.

Data visualization and descriptive statistics.

Data splitting into training (80%) and test sets (20%) in an effective way.
HARDWARE REQUIREMENTS

 Processor- Intel (R) Core (TM) i3-4200U

 CPU - 1.6GHz

 RAM:4 GB

 Hard Disk: 500 GB.


SOFTWARE REQUIREMENTS

 Operating System - Windows 10


 Server - Apache Tomcat
 Front End - HTML, CSS, JS
 Back End - JSP
 Database - MYSQL server 5.0

You might also like