Cystoscopic Image Classification by Unsupervised Feature Learning and Fusion of Classifiers

Abstract

Cystoscopy imaging is highly recommended for the early diagnosis of bladder cancer, which is the ninth most common cancer in the world. This study presents an intelligent method for classifying cystoscopy images of bladder. In the proposed method, a pre-trained convolutional neural network (CNN) is employed to extract high level semantic features. Then, the number of features is reduced using Principal Component Analysis (PCA) followed by Linear Discriminant Analysis (LDA) to avoid curse of dimensionality issue. In the classification phase, an ensemble classifier is constructed by combining Support Vector Machine (SVM), Logistic Regression, Random Forest, and Gaussian Naïve Bayes using weighted majority vote. The proposed method is evaluated on 720 cystoscopy images collected in a medical center. Next, the suggested method is categorized into four different classes including bloody urine, benign masses, malignant masses, and normal cases. The results of the experiments indicated that the presented work achieved an accuracy of 69.02 ± 0.19, which outperformed other competing methods.