Segmentation of laser induced retinal lesions using deep learning (December 2021).

Autor: Gil EM; Department of Biomedical Engineering, Texas A&M University, College Station, Texas, USA.; SAIC, JBSA Fort Sam, Houston, Texas, USA., Keppler M; Department of Biomedical Engineering, Texas A&M University, College Station, Texas, USA.; SAIC, JBSA Fort Sam, Houston, Texas, USA., Boretsky A; SAIC, JBSA Fort Sam, Houston, Texas, USA., Yakovlev VV; Department of Biomedical Engineering, Texas A&M University, College Station, Texas, USA., Bixler JN; Air Force Research Laboratory, JBSA Fort Sam, Houston, Texas, USA.
Jazyk: angličtina
Zdroj: Lasers in surgery and medicine [Lasers Surg Med] 2022 Oct; Vol. 54 (8), pp. 1130-1142. Date of Electronic Publication: 2022 Jul 03.
DOI: 10.1002/lsm.23578
Abstrakt: Objective: Detection of retinal laser lesions is necessary in both the evaluation of the extent of damage from high power laser sources, and in validating treatments involving the placement of laser lesions. However, such lesions are difficult to detect using Color Fundus cameras alone. Deep learning-based segmentation can remedy this, by highlighting potential lesions in the image.
Methods: A unique database of images collected at the Air Force Research Laboratory over the past 30 years was used to train deep learning models for classifying images with lesions and for subsequent segmentation. We investigate whether transferring weights from models that learned classification would improve performance of the segmentation models. We use Pearson's correlation coefficient between the initial and final training phases to reveal how the networks are transferring features.
Results: The segmentation models are able to effectively segment a broad range of lesions and imaging conditions.
Conclusion: Deep learning-based segmentation of lesions can effectively highlight laser lesions, making this a useful tool for aiding clinicians.
(© 2022 Wiley Periodicals LLC.)
Databáze: MEDLINE
Externí odkaz: