Skip to content
Open access

AI-Based Visual Prognosis in Full-Thickness Macular Hole Surgery Using the ILM Flap Technique

Jul 2026 · Beyoglu Eye Journal · Vol 11, pp. 160 - 168 · 0 citations · 28 references
Medicine

TL;DR

The AI platform showed clinically relevant predictive performance for preoperative and later postoperative visual outcomes after FTMH surgery, particularly at 6 and 12 months, but was less reliable in the early postoperative period.

Abstract

Objectives This study evaluated the clinical utility of a multimodal large language model-based artificial intelligence (AI) model in predicting postoperative visual outcomes following full-thickness macular hole (FTMH) surgery using the inverted internal limiting membrane (ILM) flap technique. Methods A retrospective analysis was conducted on 45 patients who underwent pars plana vitrectomy for FTMH at a tertiary eye care center between January 2021 and December 2023. Preoperative optical coherence tomography (OCT) images, demographic data, and clinical parameters were analyzed using the AI model to predict best-corrected visual acuity (BCVA) at four postoperative time points. The predicted BCVA values were then compared with actual clinical outcomes. Results Preoperatively, the mean AI-predicted BCVA was 1.13±0.20 logMAR compared with the actual value of 1.24±0.33 logMAR (p=0.192). At 6 months, the predicted BCVA was 0.65±0.20 logMAR versus the actual value of 0.67±0.25 logMAR (p=0.528), and at 12 months, it was 0.47±0.17 logMAR versus 0.55±0.27 logMAR (p=0.155). However, at 7 days postoperatively, the model significantly overestimated visual impairment, predicting 1.37±0.28 logMAR versus the actual value of 1.07±0.33 logMAR (p<0.001). Spearman correlation analysis showed the strongest association between AI-predicted and actual BCVA at 6 months (rs=0.5885, p<0.001), with a moderate correlation at 12 months (rs=0.4156, p=0.005), a weak correlation preoperatively (rs=0.3029, p=0.043), and no significant correlation at 7 days (rs=0.1949, p=0.199). Conclusion These findings demonstrate the model’s potential as a supportive tool for visual outcome prediction after FTMH surgery. The AI platform showed clinically relevant predictive performance for preoperative and later postoperative visual outcomes after FTMH surgery, particularly at 6 and 12 months, but was less reliable in the early postoperative period.

Read PDF

Similar papers

Open access Aug 2026

Clinical and morphological predictors of visual outcomes following inverted internal limiting membrane flap surgery for large macular holes

Purpose: The purpose of the study was to evaluate the anatomical and functional outcomes of the inverted internal limiting membrane (ILM) flap technique in patients with large macular holes (MHs) and identify independent prognostic factors that predict post-operative visual acuity. Methods: This retrospective study included 48 eyes of 48 patients who underwent pars plana vitrectomy and inverted ILM flap technique for idiopathic MHs ≥400 μm. Pre-operative and post-operative best-corrected visual acuity (BCVA), swept-source optical coherence tomography (SS-OCT) parameters, and various clinical factors were evaluated. Multivariate regression analysis was performed to identify independent prognostic factors for the post-operative BCVA. Results: The mean BCVA improved significantly from 1.27±0.41 logMAR preoperatively to 0.62±0.37 logMAR at 6 months postoperatively (p<0.001). U-shaped closure was associated with better BCVA than V-and W-shaped closures (p=0.001). Patients with symptom duration ≤6 months and complete post-operative restoration of the external limiting membrane (ELM) and ellipsoid zone demonstrated significantly better BCVA (p<0.01). Pre-operative BCVA (β=0.637, p<0.001) and MH index (MHI) (β=−0.261, p=0.029) were significant independent determinants of post-operative BCVA. Conclusion: The inverted ILM flap technique provides high anatomical closure rates and significant visual improvement in large MH surgery. Pre-operative BCVA and MHI were significant independent prognostic factors for post-operative visual outcomes. Structural SS-OCT parameters can be clinically useful in predicting post-operative functional outcomes

Güzide Akçay, Dilber Celik Yaprak, U. Kıvrak et al. · 0 citations
Open access Aug 2026

Anatomical and Functional Outcomes of Macular Hole Surgery According to Internal Limiting Membrane Flap Size

In large macular holes, an extended ILM flap of approximately 1.5–3 DD improved anatomical stability and visual recovery, representing an effective surgical strategy, suggesting flap size modification may not be necessary for optimal results.

Na Gyung Kim, Myung In Yeom, Jung Min Park · 0 citations
Open access 2026

Clinical and Surgical Predictors of Visual and Anatomical Outcomes Following Pars Plana Vitrectomy for Rhegmatogenous Retinal Detachment

The findings underscore the importance of preserving macular attachment, minimizing surgical delay, and recognizing PVR as a key determinant of both functional recovery and recurrence in optimizing outcomes after PPV for RRD.

Güzide Akçay, U. Kıvrak, I. Karagöz et al. · 0 citations
Open access Jul 2026

Postoperative OCT Microstructural Changes as Predictors of Visual Prognosis Following Epiretinal Membrane Peeling: A Retrospective Study.

AIM Idiopathic macular epiretinal membrane (iERM) induces traction and deformation of the macular retina, significantly impairing function and quality of life. This study aimed to evaluate the predictive value of early postoperative optical coherence tomography (OCT) microstructural changes for visual prognosis in patients undergoing vitrectomy with membrane peeling for iERM. METHODS A total of 110 patients diagnosed with iERM at Huzhou Aier Eye Hospital between January 2021 and July 2024 were retrospectively enrolled. According to OCT findings, patients were divided into stage I (n = 38), stage II (n = 52), and stage III (n = 20). All patients underwent 23-gauge minimally invasive vitrectomy combined with indocyanine green staining and peeling of the epiretinal membrane and internal limiting membrane. Best-corrected visual acuity (BCVA), central macular thickness (CMT), subfoveal choroidal thickness (SFCT), choroidal thickness at 1000 μm from the nasal side of the macular fovea (NFCT), and choroidal thickness at 1000 μm from the temporal side of the macular fovea (TFCT) were assessed before surgery and at 1 and 3 months postoperatively. Repeated-measures analysis of variance was used to compare intergroup differences over time, and multivariate logistic regression was performed to identify indicators associated with a favorable visual prognosis. RESULTS Significant gradient differences were observed among the three groups in CMT, SFCT, NFCT, TFCT, and BCVA, with stage I patients exhibiting significantly lower values than those in stage II and stage III (all p < 0.05). During postoperative follow-up, OCT parameters and visual acuity improved in all groups, with significantly greater improvement in stage I patients compared with stage II or stage III (all p < 0.05). Repeated-measures analysis demonstrated a significant time × group interaction effect for CMT (F = 2.334, p = 0.032), with a greater rate of CMT reduction in stage I patients. At 6 months postoperatively, 71 patients (64.55%) achieved a favorable visual prognosis. Multivariate logistic regression analysis indicated that lower CMT at 1 month postoperatively (OR = 0.98, 95% CI: 0.97-0.99, p < 0.05) was significantly associated with favorable visual outcomes at 6 months. CONCLUSIONS Postoperative anatomical and functional recovery in iERM patients is stage-dependent. Lower CMT at 1 month after surgery is significantly associated with favorable visual outcomes at 6 months, suggesting its potential value as an early postoperative prognostic indicator. These findings underscore the importance of incorporating early postoperative OCT assessment into clinical follow-up to facilitate individualized prognosis and patient counseling.

Weifeng Min, Yan-chun Zhu, Xinxin Li · 0 citations