A HYBRID DEEP LEARNING ALGORITHMS FOR DIABETES MELLITUS PREDICTION USING THERMAL FOOT IMAGES
European Journal of Molecular & Clinical Medicine,
2020, Volume 7, Issue 11, Pages 5176-5183
AbstractDiabetes mellitus, frequently known as diabetes, is a disease that affects a vast majority of people globally. Diabetes cannot be cured; it can only be kept under control. In this paper, diabetes is diagnosed by the analysis of foot Variability obtained from thermal images. We employed deep learning networks of the Convolutional neural network (CNN) and CNN-SVM (SVM-Support Vector Machine) combination to mechanically sense the anomaly. Unlike the conventional analysis methods so far followed, deep learning techniques do not require any feature extraction. We performed classiﬁcation splitting the database into separate training and testing data. The maximum accuracy obtained for test data is 97.9% using CVM(Integration of Convolutional Neural networks with Support vector machines). Using CNN+SVM gave an of 93.6% of sensitivity while CNN-SVM combination gave the maximum specificity of 90.3%. As per our best knowledge, this is the ﬁrst paper in which deep learning techniques are employed in distinguishing diabetes and normal. The accuracy obtained using cross-validation is the maximum value achieved so far for the automated detection of diabetes using thermal footpaths.
- Article View: 156
- PDF Download: 465