Presentation
16 March 2023 Glomerulus quantification with deep learning based on novel multi-modal label-free quantitative phase imaging from a near-infrared (Conference Presentation)
Hyewon Cho, Nurbolat Aimakov, Inwoo Park, Myeonghoon Choi, Yerim Kim, Geosong Na, Sunghoon Lim, Woonggyu Jung
Author Affiliations +
Proceedings Volume PC12389, Quantitative Phase Imaging IX; PC123890A (2023) https://doi.org/10.1117/12.2651095
Event: SPIE BiOS, 2023, San Francisco, California, United States
Abstract
A novel multi-modal label-free imaging system is proposed for histopathology, which provides uniformly reconstructed virtual-stained brightfield images and corresponding QPI images. The system was tested on urinal histopathology, to detect and segment glomerulus. From each modality, over 90% of IoU scores were obtained and accelerated performance was obtained through multi-modal learning. Briefly, histopathology quantification with label-free samples is a feasible method via the proposed novel system.
Conference Presentation
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Hyewon Cho, Nurbolat Aimakov, Inwoo Park, Myeonghoon Choi, Yerim Kim, Geosong Na, Sunghoon Lim, and Woonggyu Jung "Glomerulus quantification with deep learning based on novel multi-modal label-free quantitative phase imaging from a near-infrared (Conference Presentation)", Proc. SPIE PC12389, Quantitative Phase Imaging IX, PC123890A (16 March 2023); https://doi.org/10.1117/12.2651095
Advertisement
Advertisement
KEYWORDS
Imaging systems

Phase imaging

Light sources

Near infrared

Image registration

Image segmentation

Light emitting diodes

Back to Top