Cargando…
Automated Quantitative Analysis of Anterior Segment Inflammation Using Swept-Source Anterior Segment Optical Coherence Tomography: A Pilot Study
Background: The aim of this study is to develop an automated evaluation of anterior chamber (AC) cells in uveitis using anterior segment (AS) optical coherence tomography (OCT) images. Methods: We analyzed AS swept-source (SS)-OCT (CASIA 2) images of 31 patients (51 eyes) with uveitis using image an...
Autores principales: | , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9689595/ https://www.ncbi.nlm.nih.gov/pubmed/36359546 http://dx.doi.org/10.3390/diagnostics12112703 |
_version_ | 1784836574977982464 |
---|---|
author | Keino, Hiroshi Aman, Takuto Furuya, Ryota Nakayama, Makiko Okada, Annabelle A. Sunayama, Wataru Hatanaka, Yuji |
author_facet | Keino, Hiroshi Aman, Takuto Furuya, Ryota Nakayama, Makiko Okada, Annabelle A. Sunayama, Wataru Hatanaka, Yuji |
author_sort | Keino, Hiroshi |
collection | PubMed |
description | Background: The aim of this study is to develop an automated evaluation of anterior chamber (AC) cells in uveitis using anterior segment (AS) optical coherence tomography (OCT) images. Methods: We analyzed AS swept-source (SS)-OCT (CASIA 2) images of 31 patients (51 eyes) with uveitis using image analysis software (Python). An automated algorithm was developed to detect cellular spots corresponding to hyper-reflective spots in the AC, and the correlation with Standardization of Uveitis Nomenclature (SUN) grading AC cells score was evaluated. The approximated AC grading value was calculated based on the logarithmic approximation curve between the number of cellular spots and the SUN grading score. Results: Among 51 eyes, cellular spots were automatically segmented in 48 eyes, whereas three eyes (all SUN grading AC cells score: 4+) with severe fibrin formation in the AC were removed by the automated algorithm. The AC cellular spots increased with an increasing SUN grading score (p < 0.001). The 48 eyes were split into training data (26 eyes) and test data (22 eyes). There was a significant correlation between the SUN grading score and the number of cellular spots in 26 eyes (rho: 0.843, p < 0.001). There was a significant correlation between the SUN grading score and the approximated grading value of 22 eyes based on the logarithmic approximation curve (rho: 0.774, p < 0.001). Leave-one-out cross-validation analysis demonstrated a significant correlation between the SUN grading score and the approximated grading value of 48 eyes (rho: 0.748, p < 0.001). Conclusions: This automated anterior AC cell analysis using AS SS-OCT showed a significant correlation with clinical SUN grading scores and provided SUN AC grading values as a continuous variable. Our findings suggest that automated grading of AC cells could improve the accuracy of a quantitative assessment of AC inflammation using AS-OCT images and allow the objective and rapid evaluation of anterior segment inflammation in uveitis. Further investigations on a large scale are required to validate this quantitative measurement of anterior segment inflammation in uveitic eyes. |
format | Online Article Text |
id | pubmed-9689595 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-96895952022-11-25 Automated Quantitative Analysis of Anterior Segment Inflammation Using Swept-Source Anterior Segment Optical Coherence Tomography: A Pilot Study Keino, Hiroshi Aman, Takuto Furuya, Ryota Nakayama, Makiko Okada, Annabelle A. Sunayama, Wataru Hatanaka, Yuji Diagnostics (Basel) Article Background: The aim of this study is to develop an automated evaluation of anterior chamber (AC) cells in uveitis using anterior segment (AS) optical coherence tomography (OCT) images. Methods: We analyzed AS swept-source (SS)-OCT (CASIA 2) images of 31 patients (51 eyes) with uveitis using image analysis software (Python). An automated algorithm was developed to detect cellular spots corresponding to hyper-reflective spots in the AC, and the correlation with Standardization of Uveitis Nomenclature (SUN) grading AC cells score was evaluated. The approximated AC grading value was calculated based on the logarithmic approximation curve between the number of cellular spots and the SUN grading score. Results: Among 51 eyes, cellular spots were automatically segmented in 48 eyes, whereas three eyes (all SUN grading AC cells score: 4+) with severe fibrin formation in the AC were removed by the automated algorithm. The AC cellular spots increased with an increasing SUN grading score (p < 0.001). The 48 eyes were split into training data (26 eyes) and test data (22 eyes). There was a significant correlation between the SUN grading score and the number of cellular spots in 26 eyes (rho: 0.843, p < 0.001). There was a significant correlation between the SUN grading score and the approximated grading value of 22 eyes based on the logarithmic approximation curve (rho: 0.774, p < 0.001). Leave-one-out cross-validation analysis demonstrated a significant correlation between the SUN grading score and the approximated grading value of 48 eyes (rho: 0.748, p < 0.001). Conclusions: This automated anterior AC cell analysis using AS SS-OCT showed a significant correlation with clinical SUN grading scores and provided SUN AC grading values as a continuous variable. Our findings suggest that automated grading of AC cells could improve the accuracy of a quantitative assessment of AC inflammation using AS-OCT images and allow the objective and rapid evaluation of anterior segment inflammation in uveitis. Further investigations on a large scale are required to validate this quantitative measurement of anterior segment inflammation in uveitic eyes. MDPI 2022-11-05 /pmc/articles/PMC9689595/ /pubmed/36359546 http://dx.doi.org/10.3390/diagnostics12112703 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Keino, Hiroshi Aman, Takuto Furuya, Ryota Nakayama, Makiko Okada, Annabelle A. Sunayama, Wataru Hatanaka, Yuji Automated Quantitative Analysis of Anterior Segment Inflammation Using Swept-Source Anterior Segment Optical Coherence Tomography: A Pilot Study |
title | Automated Quantitative Analysis of Anterior Segment Inflammation Using Swept-Source Anterior Segment Optical Coherence Tomography: A Pilot Study |
title_full | Automated Quantitative Analysis of Anterior Segment Inflammation Using Swept-Source Anterior Segment Optical Coherence Tomography: A Pilot Study |
title_fullStr | Automated Quantitative Analysis of Anterior Segment Inflammation Using Swept-Source Anterior Segment Optical Coherence Tomography: A Pilot Study |
title_full_unstemmed | Automated Quantitative Analysis of Anterior Segment Inflammation Using Swept-Source Anterior Segment Optical Coherence Tomography: A Pilot Study |
title_short | Automated Quantitative Analysis of Anterior Segment Inflammation Using Swept-Source Anterior Segment Optical Coherence Tomography: A Pilot Study |
title_sort | automated quantitative analysis of anterior segment inflammation using swept-source anterior segment optical coherence tomography: a pilot study |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9689595/ https://www.ncbi.nlm.nih.gov/pubmed/36359546 http://dx.doi.org/10.3390/diagnostics12112703 |
work_keys_str_mv | AT keinohiroshi automatedquantitativeanalysisofanteriorsegmentinflammationusingsweptsourceanteriorsegmentopticalcoherencetomographyapilotstudy AT amantakuto automatedquantitativeanalysisofanteriorsegmentinflammationusingsweptsourceanteriorsegmentopticalcoherencetomographyapilotstudy AT furuyaryota automatedquantitativeanalysisofanteriorsegmentinflammationusingsweptsourceanteriorsegmentopticalcoherencetomographyapilotstudy AT nakayamamakiko automatedquantitativeanalysisofanteriorsegmentinflammationusingsweptsourceanteriorsegmentopticalcoherencetomographyapilotstudy AT okadaannabellea automatedquantitativeanalysisofanteriorsegmentinflammationusingsweptsourceanteriorsegmentopticalcoherencetomographyapilotstudy AT sunayamawataru automatedquantitativeanalysisofanteriorsegmentinflammationusingsweptsourceanteriorsegmentopticalcoherencetomographyapilotstudy AT hatanakayuji automatedquantitativeanalysisofanteriorsegmentinflammationusingsweptsourceanteriorsegmentopticalcoherencetomographyapilotstudy |