Radiomic Feature Extraction from OCT Angiography of Idiopathic Epiretinal Membranes and Correlation with Visual Acuity: A Pilot Study

IF 3.2 Q1 OPHTHALMOLOGY

Ophthalmology science Pub Date : 2025-01-21 DOI:10.1016/j.xops.2025.100716

Maria Cristina Savastano MD , Marica Vagni MD , Matteo Mario Carlà MD , Huong Elena Tran MD , Claudia Fossataro MD , Valentina Cestrone AO , Francesco Boselli MD , Federico Giannuzzi MD , Sofia Marcelli AO , Ilaria Biagini AO , Luca Boldrini MD , Stanislao Rizzo MD

{"title":"Radiomic Feature Extraction from OCT Angiography of Idiopathic Epiretinal Membranes and Correlation with Visual Acuity: A Pilot Study","authors":"Maria Cristina Savastano MD , Marica Vagni MD , Matteo Mario Carlà MD , Huong Elena Tran MD , Claudia Fossataro MD , Valentina Cestrone AO , Francesco Boselli MD , Federico Giannuzzi MD , Sofia Marcelli AO , Ilaria Biagini AO , Luca Boldrini MD , Stanislao Rizzo MD","doi":"10.1016/j.xops.2025.100716","DOIUrl":null,"url":null,"abstract":"<div><h3>Purpose</h3><div>To explore the correlation between radiomics features extracted from OCT angiography (OCTA) of epiretinal membranes (ERMs) and baseline best-corrected visual acuity (BCVA).</div></div><div><h3>Design</h3><div>Retrospective observational monocentric study.</div></div><div><h3>Participants</h3><div>Eighty-three eyes affected by idiopathic ERMs, categorized into low (≤70 letters) and high (70 letters) BCVA groups.</div></div><div><h3>Methods</h3><div>The central 3 × 3 mm<sup>2</sup> crop of structural and vascular en-face OCTA scans of superficial and deep retina slab, and choriocapillaris of each eye was selected. PyRadiomics was used to extract 86 features belonging to 2 different families: intensity-based statistical features describing the gray-level distribution, and textural features capturing the spatial arrangement of pixels. By employing a greedy strategy, 4 radiomic features were selected to build the final logistic regression model. The ability of the model to discriminate between low and high baseline BCVA was quantified in terms of area under the receiver operating characteristics curve (AUC).</div></div><div><h3>Main Outcome Measures</h3><div>The 4 selected informative radiomic features were as follows: the difference average (glcm_DifferenceAverage), quantifying the average difference in gray-level between neighboring pixels; the informational measure of correlation (glcm_Imc1), giving information about the spatial correlation of pixel intensities inside the image; the long run low gray-level emphasis (glrlm_LongRunLowGrayLevelEmphasis), highlighting long segments of low gray-level values within the image; and the large area emphasis (glszm_LargeAreaEmphasis), which quantifies the tendency for larger zones of uniform intensity to occur.</div></div><div><h3>Results</h3><div>No features exhibited a statistically significant difference between low and high BCVA values for the superficial and deep retinal slabs. Conversely, in the choriocapillaris layer, the glcm_DifferenceAverage and glcm_Imc1 features were significantly higher in the high BCVA group (<em>P</em> = 0.047), whereas higher values for the glrlm_LongRunLowGrayLevelEmphasis and glszm_LargeAreaEmphasis were associated with the low BCVA group (<em>P</em> = 0.047). Overall, these radiomic features predicted BCVA with an AUC (95% confidence interval) of 0.74 (0.63–0.85) and sensitivity/specificity of 0.67/0.75. During the cross-validation, the metrics remained stable.</div></div><div><h3>Conclusions</h3><div>Radiomics features of the choriocapillaris in idiopathic ERMs showed a correlation with BCVA, with lower structural complexity and higher homogeneity, together with the presence of homogeneous areas with low-intensity pixel values, reflecting flow voids due to reduced microvascular perfusion, and were correlated with lower visual acuity.</div></div><div><h3>Financial Disclosure(s)</h3><div>The author(s) have no proprietary or commercial interest in any materials discussed in this article.</div></div>","PeriodicalId":74363,"journal":{"name":"Ophthalmology science","volume":"5 3","pages":"Article 100716"},"PeriodicalIF":3.2000,"publicationDate":"2025-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ophthalmology science","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2666914525000144","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"OPHTHALMOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Purpose

To explore the correlation between radiomics features extracted from OCT angiography (OCTA) of epiretinal membranes (ERMs) and baseline best-corrected visual acuity (BCVA).

Design

Retrospective observational monocentric study.

Participants

Eighty-three eyes affected by idiopathic ERMs, categorized into low (≤70 letters) and high (70 letters) BCVA groups.

Methods

The central 3 × 3 mm² crop of structural and vascular en-face OCTA scans of superficial and deep retina slab, and choriocapillaris of each eye was selected. PyRadiomics was used to extract 86 features belonging to 2 different families: intensity-based statistical features describing the gray-level distribution, and textural features capturing the spatial arrangement of pixels. By employing a greedy strategy, 4 radiomic features were selected to build the final logistic regression model. The ability of the model to discriminate between low and high baseline BCVA was quantified in terms of area under the receiver operating characteristics curve (AUC).

Main Outcome Measures

The 4 selected informative radiomic features were as follows: the difference average (glcm_DifferenceAverage), quantifying the average difference in gray-level between neighboring pixels; the informational measure of correlation (glcm_Imc1), giving information about the spatial correlation of pixel intensities inside the image; the long run low gray-level emphasis (glrlm_LongRunLowGrayLevelEmphasis), highlighting long segments of low gray-level values within the image; and the large area emphasis (glszm_LargeAreaEmphasis), which quantifies the tendency for larger zones of uniform intensity to occur.

Results

No features exhibited a statistically significant difference between low and high BCVA values for the superficial and deep retinal slabs. Conversely, in the choriocapillaris layer, the glcm_DifferenceAverage and glcm_Imc1 features were significantly higher in the high BCVA group (P = 0.047), whereas higher values for the glrlm_LongRunLowGrayLevelEmphasis and glszm_LargeAreaEmphasis were associated with the low BCVA group (P = 0.047). Overall, these radiomic features predicted BCVA with an AUC (95% confidence interval) of 0.74 (0.63–0.85) and sensitivity/specificity of 0.67/0.75. During the cross-validation, the metrics remained stable.

Conclusions

Radiomics features of the choriocapillaris in idiopathic ERMs showed a correlation with BCVA, with lower structural complexity and higher homogeneity, together with the presence of homogeneous areas with low-intensity pixel values, reflecting flow voids due to reduced microvascular perfusion, and were correlated with lower visual acuity.

Financial Disclosure(s)

The author(s) have no proprietary or commercial interest in any materials discussed in this article.

查看原文本刊更多论文

特发性视网膜上膜的OCT血管造影放射特征提取及其与视力的相关性：一项初步研究

目的探讨视网膜前膜（ERMs） OCT血管造影（OCTA）放射组学特征与基线最佳矫正视力（BCVA）的相关性。设计回顾性观察性单中心研究。参与者有83只眼睛受到特发性erm的影响，分为低（≤70个字母）和高（70个字母）BCVA组。方法选取每只眼的浅、深视网膜板和绒毛膜毛细血管的中央3 × 3 mm2的结构和血管面OCTA扫描。使用PyRadiomics提取了86个特征，分别属于2个不同的科：基于强度的描述灰度分布的统计特征和捕获像素空间排列的纹理特征。采用贪心策略，选取4个放射特征构建最终的逻辑回归模型。模型区分低基线和高基线BCVA的能力通过受试者工作特征曲线下面积（AUC）进行量化。主要观察指标选取的4个信息放射学特征为：差分平均值（glcm_DifferenceAverage），量化相邻像素间灰度水平的平均差异；相关性的信息度量（glcm_Imc1），给出图像内像素强度的空间相关性信息；长期低灰度强调（glrlm_LongRunLowGrayLevelEmphasis），突出显示图像中低灰度值的长段；以及大面积强调（glszm_LargeAreaEmphasis），它量化了出现较大均匀强度区域的趋势。结果浅、深视网膜板低、高BCVA值差异无统计学意义。相反，在绒毛膜层，高BCVA组glcm_DifferenceAverage和glcm_Imc1特征显著较高（P = 0.047），而glrlm_LongRunLowGrayLevelEmphasis和glszm_LargeAreaEmphasis值较高与低BCVA组相关（P = 0.047）。总体而言，这些放射学特征预测BCVA的AUC（95%置信区间）为0.74(0.63-0.85)，敏感性/特异性为0.67/0.75。在交叉验证期间，度量保持稳定。结论特发性erm的绒毛膜毛细血管放射组学特征与BCVA相关，具有较低的结构复杂性和较高的均匀性，同时存在低强度像素值的均匀区域，反映微血管灌注减少导致的血流空洞，与视力降低相关。财务披露作者在本文中讨论的任何材料中没有专有或商业利益。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊