Computer-Aided Detection and Diagnosis of Thyroid Nodules Using Machine and Deep Learning Classification Algorithms

Abstract

This paper proposes a computer-aided methodology for detecting and segmenting the tumor regions in ultrasound thyroid images using machine and deep learning algorithms. This proposed tumor detection methodology uses Kirsch’s edge detector for enhancing the edge region pixels in thyroid image and then Dual Tree Contourlet Transform (DTCT) was applied on the enhanced image for obtaining the coefficients. Then, features are computed from this transformed thyroid image and these features are trained and classified using the Co-Active Adaptive Neuro Expert System (CANFES) classifier. Then, the morphological segmentation method is applied on the abnormal thyroid image to segment the tumor regions. Finally, the Convolutional Neural Network (CNN) algorithm is applied on the segmented tumor regions for diagnosing them into mild, moderate and severe.

KEYWORDS:

• Ultrasound image
• Tumor
• Thyroid
• Features
• Diagnosis

Additional information

Notes on contributors

B. Shankarlal

B Shankarlal is assistant professor in the Department of Electronics and Communication Engineering at Perunthalaivar Kamarajar Institute of Engineering and Technology (Affiliated to Pondicherry University), Karaikal, India. He received his BE degree in electronics and communication engineering from Periyar University, India, in 2004 and ME degree in applied electronics from the Anna University, India, in 2006. He has 14 years of experience in teaching. He is currently pursuing his PhD degree from Annamalai University, India and his areas of interest are digital image processing, medical image processing, computer vision and image processing and optimization techniques.

P. D. Sathya

P D Sathya is assistant professor in the Department of Electronics and Communication Engineering at Annamalai University, India. She obtained BE (Electronics and Communication), ME (Applied Electronics) and PhD degrees from Periyar University, Anna University and Annamalai University in the years 2003, 2005 and 2012, respectively. She has 15 years of experience in teaching and research and development with specialization in signal processing, image and video processing and communication fields. She has published more than 40 research papers in reputed international journals including Elsevier and Inderscience, has presented more than 30 papers in various international conferences. She has guided one PhD scholar and 6 research scholars are doing research under her guidance. She has been a part of various seminars, paper presentations, research paper reviews, and conferences as a convener and a session chair, a guest editor in journals. Her research interests include signal processing, image and video processing and optimization techniques applied to various image processing applications. Email: [email protected]

V. P. Sakthivel

V P Sakthivel is assistant professor in the Department of Electrical and Electronics Engineering (EEE) at Government College of Engineering, Dharmapuri, India. He obtained BE (EEE), ME (Power Systems) and PhD degree from Madras University, Anna University and Annamalai University in the years 2001, 2004 and 2012, respectively. He has 16 years of experience in teaching and research and development with specialization in electrical machine design, power systems and applications of heuristic algorithms for power system optimization image fields. He is now guiding five students in their PhD program. He has published more than 50 research papers in reputed international journals including IET, Elsevier, Taylor and Francis and Compel, has presented more than 40 papers in various international conferences. He has been a part of various seminars, paper presentations, research paper reviews, and conferences as a convener and a session chair and a guest editor in journals. Email: [email protected]