Prevalence and Associated Factors of Segmentation Errors in the Peripapillary Retinal Nerve Fiber Layer and Macular Ganglion Cell Complex in Spectral-domain Optical Coherence Tomography Images

Miki, Atsuya; Kumoi, Miho; Usui, Shinichi; Endo, Tamio; Kawashima, Rumi; Morimoto, Takeshi; Matsushita, Kazunobu; Fujikado, Takashi; Nishida, Kohji

doi:10.1097/ijg.0000000000000771

Cited by 27 publications

(30 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… 13 – 22 This may be due to our broad inclusion of several different types of artifacts, as opposed to only segmentation errors, which were the focus of other papers. 15 , 17 , 19 , 20 …”

Section: Discussionmentioning

confidence: 99%

“…However, studies thus far have shown that gender is not associated with artifact frequency, 13 , 44 and there is conflicting evidence for whether age is a significant factor. 13 , 20 , 24 , 44 , 45 Second, our patient population did not include subjects with pre-perimetric glaucoma, for whom SD-OCT measurements play an important role in management. Nevertheless, we believe that these patients are likely captured within the range of optic nerve and peripapillary RNFL characteristics defined by glaucoma and normal subjects.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Artifact Rates for 2D Retinal Nerve Fiber Layer Thickness Versus 3D Neuroretinal Rim Thickness Using Spectral-Domain Optical Coherence Tomography

Park

Tsikata

Lee

et al. 2020

Trans. Vis. Sci. Tech.

View full text Add to dashboard Cite

Purpose To compare the rates of clinically significant artifacts for two-dimensional peripapillary retinal nerve fiber layer (RNFL) thickness versus three-dimensional (3D) neuroretinal rim thickness using spectral-domain optical coherence tomography (SD-OCT). Methods Only one eye per patient was used for analysis of 120 glaucoma patients and 114 normal patients. For RNFL scans and optic nerve scans, 15 artifact types were calculated per B-scan and per eye. Neuroretinal rim tissue was quantified by the minimum distance band (MDB). Global MDB neuroretinal rim thicknesses were calculated before and after manual deletion of B-scans with artifacts and subsequent automated interpolation. A clinically significant artifact was defined as one requiring manual correction or repeat scanning. Results Among glaucomatous eyes, artifact rates per B-scan were significantly more common in RNFL scans (61.7%, 74 of 120) compared to B-scans in neuroretinal rim volume scans (20.9%, 1423 of 6820) (95% confidence interval [CI], 31.6–50.0; P < 0.0001). For clinically significant artifact rates per eye, optic nerve scans had significantly fewer artifacts (15.8% of glaucomatous eyes, 13.2% of normal eyes) compared to RNFL scans (61.7% of glaucomatous eyes, 25.4% of normal eyes) (glaucoma group: 95% CI, 34.1–57.5, P < 0.0001; normal group: 95% CI, 1.3–23.3, P = 0.03). Conclusions Compared to the most commonly used RNFL thickness scans, optic nerve volume scans less frequently require manual correction or repeat scanning to obtain accurate measurements. Translational Relevance This paper illustrates the potential for 3D OCT algorithms to improve in vivo imaging in glaucoma.

show abstract

“… 13 – 22 This may be due to our broad inclusion of several different types of artifacts, as opposed to only segmentation errors, which were the focus of other papers. 15 , 17 , 19 , 20 …”

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Artifact Rates for 2D Retinal Nerve Fiber Layer Thickness Versus 3D Neuroretinal Rim Thickness Using Spectral-Domain Optical Coherence Tomography

Park

Tsikata

Lee

et al. 2020

Trans. Vis. Sci. Tech.

View full text Add to dashboard Cite

show abstract

“…Segmentation is important for analysing the thickness of the retinal layer using OCT results. Although image quality is a critical factor for segmentation, ocular structural factors such as axial length, shape of optic disc, or tortuosity of retinal vessel also affect segmentation 3 , 10 , 11 . The superotemporal, inferonasal, inferotemporal, and temporoinferior sectors contain retinal blood vessels, which contribute to the structural variation of the parapapillary area.…”

Section: Discussionmentioning

confidence: 99%

Effect of image quality fluctuations on the repeatability of thickness measurements in swept-source optical coherence tomography

Yang

Lee

Bae

et al. 2020

Sci Rep

View full text Add to dashboard Cite

This study investigated the effect of image quality fluctuations on the repeatability of thickness measurements of the peripapillary retinal nerve fibre (PP-RNFL) and ganglion cell-inner plexiform (GC-IPL) layers using swept-source optical coherence tomography (SS-OCT). Three consecutive OCT scans each were performed on 56 healthy subject. Finally, 168 SS-OCT results were analysed. Based on the tertile values of the mean absolute difference of image quality score, all subjects were divided into the following three groups—low-(LIQD), moderate-(MIQD), and high-(HIQD) image quality score difference groups. A linear mixed model and intraclass correlation coefficients (ICCs) were used for analyses. Despite high ICC values (> 0.9), several sectors showed significant differences in the ICC values in intergroup comparisons. For LIQD-HIQD and MIQD-HIQD, most PP-RNFL sectors showed significant differences. For GC-IPL sectors, the LIQD-HIQD comparison showed significant differences in the temporosuperior ( p = 0.012), inferior ( p < .001), and temporoinferior ( p = 0.042) sectors. Significant differences existed in the average GC-IPL ( p = 0.009), nasoinferior ( p = 0.035), and inferior GC-IPL sectors ( p < .001) for MIQD-HIQD comparison. With higher image quality fluctuations, the repeatability of SS-OCT decreased in several sectors, which are considered clinically relevant in evaluating glaucoma status. Therefore, maintaining high-quality image status is essential to enhance the reliability of SS-OCT.

show abstract

“…However, artifacts during automated segmentation of the RNFL on SDOCT are not uncommon. Several studies have reported high rates of errors on segmentation of the RNFL across different SDOCT instruments 8–11,18,19 . Since most of the cohort studies in the literature that reported on SDOCT for glaucoma diagnosis excluded images with artifacts, the rate of segmentation errors could be substantially higher in routine clinical practice 8 .…”

Section: Discussionmentioning

confidence: 99%

Detecting Retinal Nerve Fibre Layer Segmentation Errors on Spectral Domain-Optical Coherence Tomography with a Deep Learning Algorithm

Jammal

Ogata

Mariottoni

et al. 2019

Sci Rep

View full text Add to dashboard Cite

In this study we developed a deep learning (DL) algorithm that detects errors in retinal never fibre layer (RNFL) segmentation on spectral-domain optical coherence tomography (SDOCT) B-scans using human grades as the reference standard. A dataset of 25,250 SDOCT B-scans reviewed for segmentation errors by human graders was randomly divided into validation plus training (50%) and test (50%) sets. The performance of the DL algorithm was evaluated in the test sample by outputting a probability of having a segmentation error for each B-scan. The ability of the algorithm to detect segmentation errors was evaluated with the area under the receiver operating characteristic (ROC) curve. Mean DL probabilities of segmentation error in the test sample were 0.90 ± 0.17 vs. 0.12 ± 0.22 (P < 0.001) for scans with and without segmentation errors, respectively. The DL algorithm had an area under the ROC curve of 0.979 (95% CI: 0.974 to 0.984) and an overall accuracy of 92.4%. For the B-scans with severe segmentation errors in the test sample, the DL algorithm was 98.9% sensitive. This algorithm can help clinicians and researchers review images for artifacts in SDOCT tests in a timely manner and avoid inaccurate diagnostic interpretations.

show abstract

Prevalence and Associated Factors of Segmentation Errors in the Peripapillary Retinal Nerve Fiber Layer and Macular Ganglion Cell Complex in Spectral-domain Optical Coherence Tomography Images

Cited by 27 publications

References 20 publications

Artifact Rates for 2D Retinal Nerve Fiber Layer Thickness Versus 3D Neuroretinal Rim Thickness Using Spectral-Domain Optical Coherence Tomography

Artifact Rates for 2D Retinal Nerve Fiber Layer Thickness Versus 3D Neuroretinal Rim Thickness Using Spectral-Domain Optical Coherence Tomography

Effect of image quality fluctuations on the repeatability of thickness measurements in swept-source optical coherence tomography

Detecting Retinal Nerve Fibre Layer Segmentation Errors on Spectral Domain-Optical Coherence Tomography with a Deep Learning Algorithm

Contact Info

Product

Resources

About