Scholarlink

Background <p>To compare the detection and quantification of intraretinal microvascular abnormalities (IRMAs) and neovascularization (NV) between widefield colour fundus photography (WF CFP) and single-capture ultrawide-field swept-source OCT angiography (UWF SS-OCTA) across diabetic retinopathy (DR) severity stages, and to evaluate whether paired multimodal review enhances lesion detection and disease classification.</p> Methods <p>Eyes of different DR stages were imaged using 133° WF CFP and 150° UWF OCTA imaging modalities. Images were independently graded by masked, trained graders for the presence and quantification of NVs and IRMAs. Paired multimodal review was then performed bidirectionally for WF CFP and UWF OCTA to assess whether lesion detection could be improved by cross-referencing the corresponding image from the other modality. Intergrader agreement and inter-modality agreement during independent and paired multimodal review were assessed using intraclass correlation coefficients (ICC).</p> Results <p>A total of 120 eyes were analysed (NPDR: 64.6%; PDR: 35.4%). Under independent grading, UWF OCTA detected significantly more lesions than WF CFP across all DR stages (all <i>p</i> &lt; 0.001), with poor inter-modality agreement. Paired multimodal review significantly increased WF CFP lesion counts across all groups (all <i>p</i> &lt; 0.001), while UWF OCTA counts remained unchanged or showed no significant increase. Inter-modality agreement improved after paired review (mild–moderate NPDR IRMA ICC: 0.219; severe NPDR IRMA ICC: 0.334; PDR NV ICC: 0.726), however, UWF OCTA continued to detect significantly more lesions (all <i>p</i> &lt; 0.001). Inter-grader agreement for both NV and IRMA grading was higher for UWF OCTA than WF CFP (NV ICC: 0.974 vs. 0.694; IRMA ICC: 0.635 vs. 0.420). In an exploratory finding, 12 eyes (10%) were reclassified from NPDR to PDR following multimodal review.</p> Conclusions <p>Single-capture UWF OCTA detects significantly more IRMAs and NV than WF CFP, with superior reproducibility. Multimodal review enhances lesion recognition on WF CFP but does not eliminate inter-modality discrepancies. These findings highlight the potential of UWF OCTA to refine DR staging and influence clinical decision-making, supporting its complementary role alongside fundus photography in specialised diabetic eye care.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Comparative detection of diabetic retinopathy microvascular lesions using single-capture ultrawide-field swept-source OCT angiography versus widefield colour fundus photography

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月25日

查看全文
  • Yusra Asad,
  • Pavan Paniharam,
  • Ser Koon Goh,
  • Ayush Jayakrishnan,
  • Wenjun Song,
  • Jun Ting Yeo,
  • Ryan Zheng Hui Lee,
  • Caitlin Loh,
  • Wei Hing Seah,
  • Ryan Zi Lin Tan,
  • Lingyi Zhuang,
  • Pei Qi Lo,
  • Tai Yong Loh,
  • Juling Sia,
  • Jia Jia Mah,
  • Ying Jie Yong,
  • Nicola Gan,
  • Rupesh Agrawal,
  • Min Wang,
  • Sandy Wenting Zhou

摘要

Background

To compare the detection and quantification of intraretinal microvascular abnormalities (IRMAs) and neovascularization (NV) between widefield colour fundus photography (WF CFP) and single-capture ultrawide-field swept-source OCT angiography (UWF SS-OCTA) across diabetic retinopathy (DR) severity stages, and to evaluate whether paired multimodal review enhances lesion detection and disease classification.

Methods

Eyes of different DR stages were imaged using 133° WF CFP and 150° UWF OCTA imaging modalities. Images were independently graded by masked, trained graders for the presence and quantification of NVs and IRMAs. Paired multimodal review was then performed bidirectionally for WF CFP and UWF OCTA to assess whether lesion detection could be improved by cross-referencing the corresponding image from the other modality. Intergrader agreement and inter-modality agreement during independent and paired multimodal review were assessed using intraclass correlation coefficients (ICC).

Results

A total of 120 eyes were analysed (NPDR: 64.6%; PDR: 35.4%). Under independent grading, UWF OCTA detected significantly more lesions than WF CFP across all DR stages (all p < 0.001), with poor inter-modality agreement. Paired multimodal review significantly increased WF CFP lesion counts across all groups (all p < 0.001), while UWF OCTA counts remained unchanged or showed no significant increase. Inter-modality agreement improved after paired review (mild–moderate NPDR IRMA ICC: 0.219; severe NPDR IRMA ICC: 0.334; PDR NV ICC: 0.726), however, UWF OCTA continued to detect significantly more lesions (all p < 0.001). Inter-grader agreement for both NV and IRMA grading was higher for UWF OCTA than WF CFP (NV ICC: 0.974 vs. 0.694; IRMA ICC: 0.635 vs. 0.420). In an exploratory finding, 12 eyes (10%) were reclassified from NPDR to PDR following multimodal review.

Conclusions

Single-capture UWF OCTA detects significantly more IRMAs and NV than WF CFP, with superior reproducibility. Multimodal review enhances lesion recognition on WF CFP but does not eliminate inter-modality discrepancies. These findings highlight the potential of UWF OCTA to refine DR staging and influence clinical decision-making, supporting its complementary role alongside fundus photography in specialised diabetic eye care.