Texture-Based Preprocessing Framework with nnU-Net Model for Accurate Intracranial Artery Segmentation

Furkejuvvon:
Bibliográfalaš dieđut
Publikašuvnnas:Journal of Imaging vol. 11, no. 12 (2025), p. 438-459
Váldodahkki: Kyuseok, Kim
Eará dahkkit: Ji-Youn, Kim
Almmustuhtton:
MDPI AG
Fáttát:
Liŋkkat:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Fáddágilkorat: Lasit fáddágilkoriid
Eai fáddágilkorat, Lasit vuosttaš fáddágilkora!

MARC

LEADER 00000nab a2200000uu 4500
001 3286310140
003 UK-CbPIL
022 |a 2313-433X 
024 7 |a 10.3390/jimaging11120438  |2 doi 
035 |a 3286310140 
045 2 |b d20250101  |b d20251231 
100 1 |a Kyuseok, Kim  |u Institute of Human Convergence Health Science, Gachon University, 191, Hambakmoero, Yeonsu-gu, Incheon 21936, Republic of Korea; kskim502@gachon.ac.kr 
245 1 |a Texture-Based Preprocessing Framework with nnU-Net Model for Accurate Intracranial Artery Segmentation 
260 |b MDPI AG  |c 2025 
513 |a Journal Article 
520 3 |a Accurate intracranial artery segmentation from digital subtraction angiography (DSA) is critical for neurovascular diagnosis and intervention planning. Vascular extraction, which combines preprocessing methods and deep learning models, yields a high level of results, but limited preprocessing results constrain the improvement of results. We propose a texture-based contrast enhancement preprocessing framework integrated with the nnU-Net model to improve vessel segmentation in time-sequential DSA images. The method generates a combined feature mask by fusing local contrast, local entropy, and brightness threshold maps, which is then used as input for deep learning–based segmentation. Segmentation performance was evaluated using the DIAS dataset with various standard quantitative metrics. The proposed preprocessing significantly improved segmentation across all metrics compared to both the baseline and contrast-limited adaptive histogram equalization (CLAHE). Using nnU-Net, the method achieved a Dice Similarity Coefficient (DICE) of 0.83 ± 0.20 and an Intersection over Union (IoU) of 0.72 ± 0.14, outperforming CLAHE (DICE 0.79 ± 0.41, IoU 0.70 ± 0.23) and the baseline (DICE 0.65 ± 0.15, IoU 0.47 ± 0.20). Most notably, vessel connectivity (VC) dropped by over 65% relative to unprocessed images, indicating marked improvements in VC and topological accuracy. This study demonstrates that combining texture-based preprocessing with nnU-Net delivers robust, noise-tolerant, and clinically interpretable segmentation of intracranial arteries from DSA. 
653 |a Accuracy 
653 |a Preprocessing 
653 |a Deep learning 
653 |a Datasets 
653 |a Blood vessels 
653 |a Image segmentation 
653 |a Neural networks 
653 |a Veins & arteries 
653 |a Algorithms 
653 |a Time series 
653 |a Medical imaging 
653 |a Visualization 
653 |a Entropy 
653 |a Texture 
700 1 |a Ji-Youn, Kim  |u Department of Dental Hygiene, Gachon University, 191, Hambakmoero, Yeonsu-gu, Incheon 21936, Republic of Korea 
773 0 |t Journal of Imaging  |g vol. 11, no. 12 (2025), p. 438-459 
786 0 |d ProQuest  |t Advanced Technologies & Aerospace Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3286310140/abstract/embedded/L8HZQI7Z43R0LA5T?source=fedsrch 
856 4 0 |3 Full Text + Graphics  |u https://www.proquest.com/docview/3286310140/fulltextwithgraphics/embedded/L8HZQI7Z43R0LA5T?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3286310140/fulltextPDF/embedded/L8HZQI7Z43R0LA5T?source=fedsrch