U-Net Segmented Adjacent Angle Detection (USAAD) for Automatic Analysis of Corneal Nerve Structures
Measurement of corneal nerve tortuosity is associated with dry eye disease, diabetic retinopathy, and a range of other conditions. However, clinicians measure tortuosity on very different grading scales that are inherently subjective. Using in vivo confocal microscopy, 253 images of corneal nerves w...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-04-01
|
| Series: | Data |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5729/5/2/37 |
| _version_ | 1827718688233488384 |
|---|---|
| author | Philip Mehrgardt Seid Miad Zandavi Simon K. Poon Juno Kim Maria Markoulli Matloob Khushi |
| author_facet | Philip Mehrgardt Seid Miad Zandavi Simon K. Poon Juno Kim Maria Markoulli Matloob Khushi |
| author_sort | Philip Mehrgardt |
| collection | DOAJ |
| description | Measurement of corneal nerve tortuosity is associated with dry eye disease, diabetic retinopathy, and a range of other conditions. However, clinicians measure tortuosity on very different grading scales that are inherently subjective. Using in vivo confocal microscopy, 253 images of corneal nerves were captured and manually labelled by two researchers with tortuosity measurements ranging on a scale from 0.1 to 1.0. Tortuosity was estimated computationally by extracting a binarised nerve structure utilising a previously published method. A novel U-Net segmented adjacent angle detection (USAAD) method was developed by training a U-Net with a series of back feeding processed images and nerve structure vectorizations. Angles between all vectors and segments were measured and used for training and predicting tortuosity measured by human labelling. Despite the disagreement among clinicians on tortuosity labelling measures, the optimised grading measurement was significantly correlated with our USAAD angle measurements. We identified the nerve interval lengths that optimised the correlation of tortuosity estimates with human grading. We also show the merit of our proposed method with respect to other baseline methods that provide a single estimate of tortuosity. The real benefit of USAAD in future will be to provide comprehensive structural information about variations in nerve orientation for potential use as a clinical measure of the presence of disease and its progression. |
| first_indexed | 2024-03-10T20:28:56Z |
| format | Article |
| id | doaj.art-4bd86779b7954c52bbade9dcd98d82a9 |
| institution | Directory Open Access Journal |
| issn | 2306-5729 |
| language | English |
| last_indexed | 2024-03-10T20:28:56Z |
| publishDate | 2020-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Data |
| spelling | doaj.art-4bd86779b7954c52bbade9dcd98d82a92023-11-19T21:34:26ZengMDPI AGData2306-57292020-04-01523710.3390/data5020037U-Net Segmented Adjacent Angle Detection (USAAD) for Automatic Analysis of Corneal Nerve StructuresPhilip Mehrgardt0Seid Miad Zandavi1Simon K. Poon2Juno Kim3Maria Markoulli4Matloob Khushi5School of Computer Science, The University of Sydney, NSW 2006, AustraliaSchool of Computer Science, The University of Sydney, NSW 2006, AustraliaSchool of Computer Science, The University of Sydney, NSW 2006, AustraliaSchool of Optometry and Vision Science, University of New South Wales, Sydney, NSW 2052, AustraliaSchool of Optometry and Vision Science, University of New South Wales, Sydney, NSW 2052, AustraliaSchool of Computer Science, The University of Sydney, NSW 2006, AustraliaMeasurement of corneal nerve tortuosity is associated with dry eye disease, diabetic retinopathy, and a range of other conditions. However, clinicians measure tortuosity on very different grading scales that are inherently subjective. Using in vivo confocal microscopy, 253 images of corneal nerves were captured and manually labelled by two researchers with tortuosity measurements ranging on a scale from 0.1 to 1.0. Tortuosity was estimated computationally by extracting a binarised nerve structure utilising a previously published method. A novel U-Net segmented adjacent angle detection (USAAD) method was developed by training a U-Net with a series of back feeding processed images and nerve structure vectorizations. Angles between all vectors and segments were measured and used for training and predicting tortuosity measured by human labelling. Despite the disagreement among clinicians on tortuosity labelling measures, the optimised grading measurement was significantly correlated with our USAAD angle measurements. We identified the nerve interval lengths that optimised the correlation of tortuosity estimates with human grading. We also show the merit of our proposed method with respect to other baseline methods that provide a single estimate of tortuosity. The real benefit of USAAD in future will be to provide comprehensive structural information about variations in nerve orientation for potential use as a clinical measure of the presence of disease and its progression.https://www.mdpi.com/2306-5729/5/2/37U-Netdeep learningcorneal nerveautomatic analysistortuosity |
| spellingShingle | Philip Mehrgardt Seid Miad Zandavi Simon K. Poon Juno Kim Maria Markoulli Matloob Khushi U-Net Segmented Adjacent Angle Detection (USAAD) for Automatic Analysis of Corneal Nerve Structures Data U-Net deep learning corneal nerve automatic analysis tortuosity |
| title | U-Net Segmented Adjacent Angle Detection (USAAD) for Automatic Analysis of Corneal Nerve Structures |
| title_full | U-Net Segmented Adjacent Angle Detection (USAAD) for Automatic Analysis of Corneal Nerve Structures |
| title_fullStr | U-Net Segmented Adjacent Angle Detection (USAAD) for Automatic Analysis of Corneal Nerve Structures |
| title_full_unstemmed | U-Net Segmented Adjacent Angle Detection (USAAD) for Automatic Analysis of Corneal Nerve Structures |
| title_short | U-Net Segmented Adjacent Angle Detection (USAAD) for Automatic Analysis of Corneal Nerve Structures |
| title_sort | u net segmented adjacent angle detection usaad for automatic analysis of corneal nerve structures |
| topic | U-Net deep learning corneal nerve automatic analysis tortuosity |
| url | https://www.mdpi.com/2306-5729/5/2/37 |
| work_keys_str_mv | AT philipmehrgardt unetsegmentedadjacentangledetectionusaadforautomaticanalysisofcornealnervestructures AT seidmiadzandavi unetsegmentedadjacentangledetectionusaadforautomaticanalysisofcornealnervestructures AT simonkpoon unetsegmentedadjacentangledetectionusaadforautomaticanalysisofcornealnervestructures AT junokim unetsegmentedadjacentangledetectionusaadforautomaticanalysisofcornealnervestructures AT mariamarkoulli unetsegmentedadjacentangledetectionusaadforautomaticanalysisofcornealnervestructures AT matloobkhushi unetsegmentedadjacentangledetectionusaadforautomaticanalysisofcornealnervestructures |