Rahman, Ashiqur, Ahmed, Amr and Yue, Shigang (2017) Classification of tongue - glossitis abnormality. In: International Conference of Data Mining and Knowledge Engineering, 5th - 7th July, 2017, London, United Kingdom.
![]() | There is a more recent version of this item available. |
Full text not available from this repository.
Item Type: | Conference or Workshop contribution (Paper) |
---|---|
Item Status: | Live Archive |
Abstract
An approach to classify tongue abnormality related to Diabetes Mellitus (DM) following Western Medicine (WM) approach. Glossitis abnormality is one of the common tongue abnormalities that affects patients who suffer from Diabetes Mellitus (DM).
The novelty of the proposed approach is attributed to utilising visual signs that appear on tongue due to Glossitis abnormality causes by high blood sugar level in the human body. The test for the blood sugar level is inconvenient for some patients in rural and poor areas where medical services are minimal or may not be available at all. To screen and monitor human organ effectively, the proposed computer aided model predicts and classifies abnormality appears on the tongue or tongue surface using visual signs caused by the abnormality. The visual signs were extracted following a logically formed medical approach, which complies with Western Medicine (WM) approach. Using Random Forest classifier on the extracted visual tongue signs, from 572 tongue samples for 166 patients, the experimental results have shown promising accuracy of 95.8% for Glossitis abnormality.
Keywords: | Tongue Classification, Random Forest, Machine Learning, Western Medicine |
---|---|
Subjects: | G Mathematical and Computer Sciences > G760 Machine Learning G Mathematical and Computer Sciences > G400 Computer Science |
Divisions: | College of Science > School of Computer Science |
ID Code: | 27400 |
Deposited On: | 28 Apr 2017 08:58 |
Available Versions of this Item
- Classification of tongue - glossitis abnormality. (deposited 28 Apr 2017 08:58) [Currently Displayed]
Repository Staff Only: item control page