DATABASE AND DISEASE CLUSTERING ANALYSIS DESIGN BASED ON BANDUNG COMMUNITY HEALTH CENTRE MONTHLY REPORT TYPE 1 FOR DEVELOPING A GEOGRAPHIC INFORMATION SYSTEM
Recording reports is one of the main activities that is done by the community health centre. One of the results of this activity is the MR-1 that contains aggregate data on disease cases based on age ranges and type of the case. This report is then submitted to the Regional Health Office (RHO) to...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/51144 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Recording reports is one of the main activities that is done by the community health
centre. One of the results of this activity is the MR-1 that contains aggregate data
on disease cases based on age ranges and type of the case. This report is then
submitted to the Regional Health Office (RHO) to be processed into information
that will support decision-making. At Bandung RHO, the storage of these reports
which are still in the form of tables is deemed inefficient in supporting decisionmaking. In an effort to increase the efficiency of the decision-making process, a GIS
that can produce a visualization of MR-1 data in the form of a map has been
developed. In addition, this GIS is also equipped with a disease clustering
subsystem to determine whether the observed cases exhibit any systematic pattern.
As a part of an information system, one of the factors that affect the performance
of this GIS is database design and implementation. So, this final project will discuss
about database and disease clustering subsystem design for GIS.
Database design consists of four stages, namely conceptual design, DBMS
selection, logical design, and physical design. This design is then implemented
using MySQL DBMS and uploaded to the JawsDB server so that the data is
accessible for GIS through cloud. The resulting database is the tested against the
database design antipatterns, black box testing, and the estimated storage
requirements are calculated. For disease clustering, the method that is used is the
SaTScan method which is based on the multiple hypothesis testing method. This
subsystem is implemented in the R programming language and uploaded to the
shiny server. This subsystem evaluation will use the controlled data and the 2019
MR-1 data as inputs. Based on the tests conducted, there are 3 out of 11 antipatterns
in the database, a range of 0,01-3,27 seconds query time, and it is known that the
storage requirement is 134 GB for 25 years. As for the disease clustering subsystem,
clusters have been formed with the log likelihood ratio as a parameter of statistical
significance for each cluster.
|
|---|