A Systematic Approach for MRI Brain Tumor Localization, and Segmentation using Deep Learning and Active Contouring

Gunasekara, Shanaka Ramesh; Kaldera, H. N. T. K.; Dissanayake, Maheshi B.

doi:10.48550/arxiv.2102.03532

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the same work, interpolation technique is used to get a high-resolution image from the low-resolution and the interpolation technique applied to enhance fusion results before segmentation and then, the threshold and the watershed segmentation methods are used sequentially to restrict the tumor region. Gunasekara et al [10] proposed deep learning architecture, to classify using deep convolutional neural network and region-based CNN is used on the classified images to localize the tumor, after that the tumor boundary is contoured for the segmentation process by using the chan-vese segmentation method. Mohan et.…”

Section: Related Workmentioning

confidence: 99%

Interpolation Method for Identification of Brain Tumor from Magnetic Resonance Images

Singh¹,

Saxena²

2023

IJEM

View full text Add to dashboard Cite

During the past years, it is observed from the literature that, identification of the brain tumor identification in human being is gaining popularity. Diagnosing any disease without manual interaction with great accuracy makes computer science research more demanding, therefore, the present work is related to identify the tumor clots in the affected patients. For this purpose, a well-known Safdarganj Hospital, New Delhi, India is consulted and 2165 Magnetic Resonance Images (MRI) of a single patient are collected through scanning, and interpolation technique of numerical method used to identify the accurate position of the brain tumor. A system model is developed and implemented by the use of Python programming language and MATLAB for the identification of affected areas in the form of a contour of a patient. The desired accuracy and specificity are evaluated using the computed results and also presented in the form of graphs.

show abstract

Section: Related Workmentioning

confidence: 99%