TY - GEN
T1 - Effect of Anterior Commisure - Posterior Commisure Reorientation for 3D Aneurysm Segmentation in the Brain
AU - Ryan, Richard
AU - Sarno, Riyanarto
AU - Suroto, Nur Setiawan
AU - Supriyanto, Muhammad Ibadurrahman Arrasyid
AU - Sihaj, Gerry
AU - Susilo, Rahadian Indarto
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - This study explores the effect of Anterior Commissure - Posterior Commissure (AC-PC) reorientation on 3D aneurysm segmentation accuracy in brain MRI (Magnetic Resonance Imaging) using the UNetR (UNet with Transformers) model. Accurate segmentation is critical in diagnosing life-threatening conditions like hemorrhagic stroke from aneurysm rupture. Given the importance of consistent imaging, AC-PC landmarks serve as valuable reference points for enhancing alignment. We conducted two preprocessing experiments: (1) reshaped to (256, 256, 256), and (2) reshaped to (256, 256, 256) with rotation based on the AC-PC line. Each experiment was evaluated by Dice score coefficient, yielding average scores of 0.22 (SD=0.19), and 0.24 (SD=0.21), respectively. While a paired t-test showed no statistically significant effect from AC-PC reorientation, the rotated data produced an approximate 9.4% improvement in segmentation accuracy.
AB - This study explores the effect of Anterior Commissure - Posterior Commissure (AC-PC) reorientation on 3D aneurysm segmentation accuracy in brain MRI (Magnetic Resonance Imaging) using the UNetR (UNet with Transformers) model. Accurate segmentation is critical in diagnosing life-threatening conditions like hemorrhagic stroke from aneurysm rupture. Given the importance of consistent imaging, AC-PC landmarks serve as valuable reference points for enhancing alignment. We conducted two preprocessing experiments: (1) reshaped to (256, 256, 256), and (2) reshaped to (256, 256, 256) with rotation based on the AC-PC line. Each experiment was evaluated by Dice score coefficient, yielding average scores of 0.22 (SD=0.19), and 0.24 (SD=0.21), respectively. While a paired t-test showed no statistically significant effect from AC-PC reorientation, the rotated data produced an approximate 9.4% improvement in segmentation accuracy.
KW - Brain Aneurysm
KW - Brain Landmarks
KW - Deep Learning
KW - Medical Image Preprocessing
KW - Segmentation
UR - https://www.scopus.com/pages/publications/105010113768
U2 - 10.1109/ICoCSETI63724.2025.11018970
DO - 10.1109/ICoCSETI63724.2025.11018970
M3 - Conference contribution
AN - SCOPUS:105010113768
T3 - ICoCSETI 2025 - International Conference on Computer Sciences, Engineering, and Technology Innovation, Proceeding
SP - 324
EP - 328
BT - ICoCSETI 2025 - International Conference on Computer Sciences, Engineering, and Technology Innovation, Proceeding
A2 - Wibowo, Ferry Wahyu
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2025 International Conference on Computer Sciences, Engineering, and Technology Innovation, ICoCSETI 2025
Y2 - 21 January 2025
ER -