Segmenting Brain Tumour Regions with Fuzzy Integrated Active Contours

Abstract

Magnetic resonance imaging used for diagnosis, localisation, and volume quantisation of tumours helps radiologists set treatment plans. Medical image segmentation is vital in detecting tumours. We propose a segmentation algorithm for identifying brain tumour regions using the fuzzy integrated active contour model. Fuzzy energy alterations define the contour and eliminate partial volume effects. We evaluated the algorithm on the BRATS 2012 and 2015 database and achieved an average dice score of 81% for the total abnormal region, including oedema, and 67% for an active tumour. This model is more robust than the classical snake methods based on the gradient.

KEYWORDS:

• Active contour
• Brain Tumour
• FCM
• Magnetic Resonance Imaging
• Multimodal
• Segmentation

Acknowledgement

Brain tumour image data used in this article were obtained from the MICCAI 2012 and 2015 Challenges on Multimodal Brain Tumour Segmentation. The challenge database contain fully anonymized images from the Cancer Imaging Atlas Archive and the BRATS 2012 challenge.

Additional information

Notes on contributors

S. Jayanthi

Jayanthi S is an associate professor in the Department of Electronics and Communication at Sakthi Mariamman Engineering College, Chennai, India. She is currently pursuing PhD degree at Anna University, Chennai, India. Her main research interests are image processing, neural networks and pattern recognition. Corresponding author. Email: [email protected]

H. Ranganathan

Ranganathan H received his BE in electronics and communication engineering from College of Engineering, Guindy, Chennai, India in 1975, ME communication systems from the same institution in 1978 and PhD from Anna university Chennai. He has over 20 years of experience in providing hardware and software solutions to various customers through India. He has around 18 years of experience in education and currently, he is professor in ECE department of Gojan School of Business and technology, Chennai. His research interests are biomedical signal processing, wireless networks, embedded systems, digital image processing. He has more than 50 technical papers to his credit published in referred International Journals or proceedings of reputed International Conferences. Email: [email protected]

M. Palanivelan

M Palanivelan, currently working as professor & head in the department of Electronics and Communication Engineering at Rajalakshmi Engineering College, Chennai, India. He has completed his PhD at Anna University, Chennai at 2015. He has published several research papers in international journals and conferences. His research interest includes peak power problems in multicarrier modulation systems, multiple antenna wireless communication systems, optical communication, internet of things and signal processing. Email: [email protected]