Have a personal or library account? Click to login
DEEP-BTS: Deep Learning based Brain Tissue Segmentation using ResU-Net Model Cover

DEEP-BTS: Deep Learning based Brain Tissue Segmentation using ResU-Net Model

Open Access
|Dec 2025

Figures & Tables

Fig. 1.

Proposed DEEP-BTS methodology.
Proposed DEEP-BTS methodology.

Fig. 2.

Pre-processing step in the proposed method.
Pre-processing step in the proposed method.

Fig. 3.

Architecture of ResU-Net.
Architecture of ResU-Net.

Fig. 4.

Experimental results of the proposed DEEP-BTS.
Experimental results of the proposed DEEP-BTS.

Fig. 5.

(a) Accuracy and (b) Loss curve of the ResU-Net.
(a) Accuracy and (b) Loss curve of the ResU-Net.

Fig. 6.

Comparison of the existing segmentation technique with ResU-Net.
Comparison of the existing segmentation technique with ResU-Net.

Fig. 7.

Segmentation comparison of standard U-Net and the proposed ResU-Net.
Segmentation comparison of standard U-Net and the proposed ResU-Net.

Performance evaluation of the DEEP-BTS_

TypesACSPREPRF1
CSF99.1297.9197.4398.7697.65
GM98.3696.7598.1497.1396.14
WM99.2598.5896.8798.8395.76
Overall98.9197.7497.4898.2496.51

Comparison of existing methods and DEEP-BTS_

AuthorsTechniquesDI

CSF [%]GM [%]WM [%]
Veluchamy, M. and Subramani, B., (2021) [23]Fuzzy C-Means878991
Yamanakkanavar, N. and Lee, B., 2020 [24]M-Net878991
Srikrishna, M., et al., (2021) [28]U-Net757982
ProposedResU-Net98.3398.0499.15

Performance comparison of the DEEP-BTS model with and without skull stripping and CSATF_

Metricswithout skull stripping without CSATFwith skull stripping without CSATFwith skull stripping with CSATF
AC97.0697.8898.91
F194.3295.9496.51
DI95.9897.6598.50
Language: English
Page range: 371 - 379
Submitted on: Feb 21, 2025
|
Accepted on: Sep 8, 2025
|
Published on: Dec 23, 2025
In partnership with: Paradigm Publishing Services
Publication frequency: Volume open

© 2025 P Sivaprakash, J Banumathi, Ashis Kumar Mishra, P Jayapriya, published by Slovak Academy of Sciences, Institute of Measurement Science
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.