Development of decision support system for identification of medically important enterobacteriaceae

Abdul Latiff, Nur Amalina (2015) Development of decision support system for identification of medically important enterobacteriaceae. Masters thesis, Universiti Sains Malaysia.

PDF - Submitted Version
Download (377kB) | Preview


Members of the family Enterobacteriaceae are the majority of gram-negative organisms identified in a clinical microbiology Laboratory. The family now has over 20 genera and more than 100 species, of which about 50 are associated with human disease. Currently, in Laboratory of Microbiology and Parasitology, of Hospital Universiti Sains Malaysia, the identification of Enterobacteriaceae is utilised routinely by conventional biochemical tests. Other than that, commercial system such as API 20E and Vitek 2 automated system are also been utilised specifically for identification of critical samples, due to its expensive cost. Identification manually by conventional method prone to human error during mixing and matching biochemical tests, which further cause misidentification, while identification using commercial methods require high cost. To overcome this problem, there is a need to develop a computerised decision support system to assist microbiologists for identification of Enterobacteriaceae. Decision support system of Enterobacteriaceae (DECIDER) were developed using free open source software, PHP and MySQL by following open source software development methodology. The newly develop system has been compared to previous method; conventional manual system, API 20E system and VITEK 2 automated system by back tested using a total of 356 positive blood culture previous record in year 2011 gathered from Laboratory of Microbiology and Parasitology. Percentage agreement was calculated. The highest percentage of complete agreement was by comparing DECIDER and Vitek 2, with 82 (87.23%) correctly identified organisms. Manual conventional system compared with DECIDER yield about 274 (76.97%) complete agreement for correctly identified organisms. Result has shown that DECIDER, identified a highly acceptable level of identification accuracy for members of the family Enterobacteriaceae. The system is simple and provides ease of use for user.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Enterobacteriaceae
Subjects: R Medicine > R Medicine (General)
Divisions: Kampus Kesihatan (Health Campus) > Pusat Pengajian Sains Perubatan (School of Medical Sciences) > Thesis
Depositing User: Mr Abdul Hadi Mohammad
Date Deposited: 16 Jul 2018 01:22
Last Modified: 16 Jul 2018 01:22

Actions (login required)

View Item View Item