Minimum visible lesion (MVL) thresholds in cynomolgus (Macaca fasicularis) retina resulting from laser exposure at wavelengths of 413-nm, 532-nm and 647-nm

Oliver, Jeffrey W.; Rockwell, Benjamin A.; Kumru, Semih S.; Clark, Clifton D.; Hodnett, Harvey M.; Noojin, Gary D.; Schuster, Kurt J.; Stolarski, David J.

doi:10.2351/1.5056746

Cited by 4 publications

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data generated has been incorporated into national and international safety standards such as the International Electrotechnical Commission 60825, and the ANSI Z 136 standards. In generating this data to support the standards, AFRL maintains a vast collection of imagery containing laser damage induced by a broad range of sources 30–32 . This study has provided a significant basis for our understanding of laser bioeffects.…”

Section: Methodsmentioning

confidence: 99%

“…In generating this data to support the standards, AFRL maintains a vast collection of imagery containing laser damage induced by a broad range of sources. [30][31][32] This study has provided a significant basis for our understanding of laser bioeffects. All retinal fundus images were acquired from Macaca fascicularis or Macaca mulatta subjects.…”

Section: Fundus Image Data Set and Preprocessingmentioning

confidence: 97%

See 1 more Smart Citation

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

Gil

Keppler

Boretsky³

et al. 2022

Lasers Surg Med

View full text Add to dashboard Cite

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.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Fundus Image Data Set and Preprocessingmentioning

confidence: 97%

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

Gil

Keppler

Boretsky³

et al. 2022

Lasers Surg Med

View full text Add to dashboard Cite

show abstract

“…Following these criteria, 43 threshold data have been excluded (from 10 studies): [10,19,57,64,65,66,67,68,69,70] and 253 threshold data published in 31 studies are found applicable to model validation: [10,11,18,19,32,33,53,54,56,57,58,59,60,61,70,71,72,73,74,55,75,76,77,78,79,80,81,82,83,84,85].…”

Section: Experimental Datamentioning

confidence: 99%

Validation of a computer model to predict laser induced thermal injury thresholds of the retina

Jean¹,

Schulmeister²

2013

International Laser Safety Conference

View full text Add to dashboard Cite

We present a computer model for predicting the median dose (ED 50) that produces an ophthamoscopically detectable lesion in the retina. It consists of an optical model (beam propagation through the eye), a reflectance model (absorption distribution within the retinal tissues), a thermal model (solving the heat equation) and a damage model (based on the Arrhenius equation). The model was validated with 253 experimental ED 50 s that cover the entire thermal regime in both macular and paramacular regions encompassing wavelengths between 413 nm and 1338 nm, pulse durations between 100 μs and 3000 s and retinal spot sizes ranging from minimum to 2 mm. These ED 50 s are matched with a mean ratio of 0.93 and a standard deviation of 31 %. The largest ratio between model prediction and experimental data was 1.7. The applicability for using the model results for risk analysis for human exposure is discussed.

show abstract

Comparison of various neural network-based models for retinal lesion analysis

Gil

Keppler

Yakovlev

et al. 2019

Optical Interactions With Tissue and Cells XXX

View full text Add to dashboard Cite

Minimum visible lesion (MVL) thresholds in cynomolgus (Macaca fasicularis) retina resulting from laser exposure at wavelengths of 413-nm, 532-nm and 647-nm

Cited by 4 publications

References 7 publications

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

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

Validation of a computer model to predict laser induced thermal injury thresholds of the retina

Comparison of various neural network-based models for retinal lesion analysis

Contact Info

Product

Resources

About