Sub‐Second Optical Coherence Tomography Angiography Protocol for Intraoral Imaging Using an Efficient Super‐Resolution Network DOI
Jinpeng Liao, Tianyu Zhang, Chunhui Li

et al.

Journal of Biophotonics, Journal Year: 2025, Volume and Issue: unknown

Published: April 20, 2025

ABSTRACT This study introduces a 200 kHz swept‐source optical coherence tomography system‐based fast angiography (OCTA) protocol for intraoral imaging by integrating an efficient Intraoral Micro‐Angiography Super‐Resolution Transformer (IMAST) model. reduces acquisition time to ~0.3 s reducing the spatial sampling resolution, thereby minimizing motion artifacts while maintaining field of view and image quality. The IMAST model utilizes transformer‐based architecture combined with convolutional operations reconstruct high‐resolution OCTA images from reduced‐resolution scans. Experimental results various sites conditions show model's robustness high performance in enhancing quality compared existing deep‐learning methods. Besides, shows advantages complexity, inference time, computational cost, underscoring its suitability clinical environments. These findings support potential our approach noninvasive oral disease diagnosis, patient discomfort facilitating early detection malignancies, thus serving as valuable tool assessment.

Language: Английский

Sub‐Second Optical Coherence Tomography Angiography Protocol for Intraoral Imaging Using an Efficient Super‐Resolution Network DOI
Jinpeng Liao, Tianyu Zhang, Chunhui Li

et al.

Journal of Biophotonics, Journal Year: 2025, Volume and Issue: unknown

Published: April 20, 2025

ABSTRACT This study introduces a 200 kHz swept‐source optical coherence tomography system‐based fast angiography (OCTA) protocol for intraoral imaging by integrating an efficient Intraoral Micro‐Angiography Super‐Resolution Transformer (IMAST) model. reduces acquisition time to ~0.3 s reducing the spatial sampling resolution, thereby minimizing motion artifacts while maintaining field of view and image quality. The IMAST model utilizes transformer‐based architecture combined with convolutional operations reconstruct high‐resolution OCTA images from reduced‐resolution scans. Experimental results various sites conditions show model's robustness high performance in enhancing quality compared existing deep‐learning methods. Besides, shows advantages complexity, inference time, computational cost, underscoring its suitability clinical environments. These findings support potential our approach noninvasive oral disease diagnosis, patient discomfort facilitating early detection malignancies, thus serving as valuable tool assessment.

Language: Английский

Citations

0