Accurate demarcation of a biased nucleus from H&E-stained follicular lymphoma tissues samples

ABSTRACT

In recent days, automated demarcation of biased nuclei of follicular lymphoma has become a standard framework in the pipeline of quantitative histopathology. It has received substantial consideration due to subjective variability between different oncologists. This difference can occasionally result in erroneous conclusions, distinct prognostic reports, and inconsistent treatment. This paper provides additional input to an oncologist to ease the prognosis process. This approach defines a local criterion fitting function in neighbourhood of each point based on image intensity. An assumption is pixel intensity will remain constant next to the point. Integration of these local neighbourhood centres leads us to define the global criterion of image segmentation. Segmentation accuracy is evaluated using region-based measures and the optimal segmentation accuracy of 98.6% is achieved using the DICE coefficient. By using locally formulated energy, execution time is also reduced in comparison to the contour algorithm. This technique is independent of the initialization.

KEYWORDS:

• Active contour model
• image segmentation
• follicular lymphoma
• local fitting energy
• centroblast

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Pranshu Saxena

Pranshu Saxena received his B. E. degree in Information Technology from the University of Pune, India in 2010 and the M. Tech. degree in Computer Science Engineering from LPU, India in 2013. Currently, he is working as an Assistant Professor in the Department of Information Technology at ABES Engineering College, India. His research interests include medical image processing, automated image segmentation, Image texture analysis, and intelligent system.

Anjali Goyal

Dr Anjali Goyal has received her Bachelor degree in Electronics in 1993 from Kurukshetra University and Master degree in Computer Applications in 1996 from Panjab University, Chandigarh. She has received her Ph.D degree from Punjab Technical University, Jalandhar, India in 2013. Presently she is working as Associate Professor in Department of Computer Application at Guru Nanak Institute of Management and Technology, Ludhiana, India. Her research interests include Content Base Image Retrieval, Digital Watermarking, Machine learning and Pattern recognition. She has a number of International journal and conference publications to her credit. She is reviewer of many reputed International journals.