A hybrid model for the identification and classification of thyroid nodules in medical ultrasound images
by Rajshree Srivastava; Pardeep Kumar
International Journal of Modelling, Identification and Control (IJMIC), Vol. 41, No. 1/2, 2022

Abstract: Ultrasonography (USG) is one of the leading diagnostic methods for accurately distinguishing the early-stage of thyroid nodules. ANN-SVM hybrid model is proposed for the identification and classification of thyroid nodules in medical ultrasound images. After feature extraction using grey level co-occurrence matrix method, two experiments are performed. In the experiment-1, five different machine learning (ML) classifiers like random forest (RF), support vector machine (SVM), decision tree (DT), artificial neural network (ANN) and K-nearest neighbour (KNN) are used for classification. While in experiment-2, the two best classifiers based on the performance are hybrid together. The proposed hybrid model has achieved 84.12% accuracy, 85.14% sensitivity and 82.95% specificity on the public dataset having 295 USG images and 90% accuracy, 91.66% sensitivity and 87.5% specificity on the local dataset having 654 thyroid USG images. It has shown an improvement of 2% to 5% in the performance evaluation in comparison with the other state-of-the-art methods.

Online publication date: Tue, 22-Nov-2022

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Modelling, Identification and Control (IJMIC):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com