Abstract
Background: Thyroid cancer is a commonly occurred cancer in all men and women around the world. This cancer can be cured if it is timely detected by proper treatment. This paper proposes a computer aided detection and screening of thyroid cancer using Adaptive Neuro Fuzzy Inference (ANFIS) classifier.
Methods: The proposed methodology has the following stages such as preprocessing, feature extraction and classification. Gaussian filter is used in the preprocessing stage to smooth the edge regions in thyroid image and further features are extracted from the preprocessed thyroid image. These extracted features are trained and classified using ANFIS classifier to classify the thyroid image into either benign or malignant.
Results: The morphological operations are further applied on the classified thyroid image to segment the cancer region. The performance of the proposed thyroid cancer detection system is analyzed in terms of sensitivity, specificity and accuracy.
Keywords: Classifier, gaussian filter, features, malignant, thyroid cancer, ANFIS.
Graphical Abstract
16
2
1
1