Rapid Bacteria Identification Using Multivariate Classifier on Diabetic Foot Infection
Identification of bacteria species on diabetic foot infection at the early stage is very crucial especially to reduce the major surgery for diabetic patients. An attempt to resolve this problem is to conduct an experiment by investigating the nature of bacterial morphology from the perspective of sensor technology. Therefore, the aim of this study is to investigate the aerobic gram positive bacteria (S. aureus, S. pyogenes) and aerobic gram negative bacteria (E. coli, P. aeruginosa, P. mirabilis and K. pneumoniae) at the early growth of bacteria at 6 hours followed by 12 hours, 18 hours to 24 hours bacteria culture in the incubator. All the bacteria were cultured in a blood agar medium by using a swabbing technique. The bacteria samples then were sniffed using Cyranose320 (e-nose) to collect the data samples and analysed using three various classifier techniques, Linear Discriminant Analysis (LDA), Probabilistic Neural Network (PNN) and k Nearest Neighbors (kNN). The preliminary results obtained in this study have shown that the e-nose was better able to identify and classify the bacteria pathogen with up to 90% accuracy using that classifier. Based on the overall classifier performance, kNN achieved higher accuracy compare to both LDA and PNN which imply that kNN is a good model to interpret the data analysis.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
Document Type: Research Article
Publication date: November 1, 2015
More about this publication?
- Journal of Medical Imaging and Health Informatics (JMIHI) is a medium to disseminate novel experimental and theoretical research results in the field of biomedicine, biology, clinical, rehabilitation engineering, medical image processing, bio-computing, D2H2, and other health related areas.
- Editorial Board
- Information for Authors
- Subscribe to this Title
- Ingenta Connect is not responsible for the content or availability of external websites