伊野田 悟

PURPOSE: To evaluate segmentation-free three-dimensional convolutional neural networks (3DCNNs) that derive visual field (VF)-shaped representations from macular optical coherence tomography (OCT) volumes (OCT-based estimated VF; OCT-VF) and to assess their performance in forecasting future Humphrey field analyzer (HFA) measurements and characterizing longitudinal variability in clinical datasets. METHODS: The 3DCNN models were trained on 129,007 heterogeneous real-world paired OCT-HFA datasets from 13,366 patients (24,313 eyes) across five institutions using 10-fold cross-validation. Forecasting accuracy for the last HFA visit was evaluated using HFA-only forecasts and hybrid models that combined HFA-anchored intercept with longitudinal slopes derived from both OCT-VF and HFA (AOS-AVG). Longitudinal variability was assessed as residual deviations from linear trends using jackknife resampling. RESULTS: The AOS-AVG yielded the lowest forecasting errors across endpoints and significantly reduced mean absolute error compared with HFA-only forecasts (pointwise thresholds: 24-2, 2.45 vs. 2.68 dB; 10-2, 2.44 vs. 2.65 dB; both P < 0.001). OCT-VF showed significantly lower residual variability than HFA (MD: 24-2, 0.79 vs. 1.13 dB; 10-2, 0.87 vs. 1.07 dB; both P < 0.001). Eyes classified as progression-positive by OCT-VF but negative by HFA exhibited greater perimetric variability and steeper HFA MD decline than true-negative eyes. CONCLUSIONS: Hybrid forecasting approaches that incorporate OCT-VF-derived slope information can improve prediction of future visual-field outcomes. The lower variability of OCT-VF representations may also enable detection of progression-related signals that remain obscured by perimetric noise in routine longitudinal monitoring. TRANSLATIONAL RELEVANCE: OCT-VF may serve as a complementary tool for precision disease monitoring alongside current automated perimetry.

PURPOSE: To determine whether excluding low-signal-strength optical coherence tomography (OCT) scans from training improves deep learning-based visual field (VF) estimation from macular OCT volumes. METHODS: We retrospectively analyzed 79,803 paired OCT and VF examinations from 8511 patients at five institutions. A three-dimensional convolutional neural network (3DCNN) estimated 24-2 and 10-2 pointwise sensitivities and mean deviation (MD) from NIDEK OCT volumes. Using patient-wise 10-fold cross-validation, we compared training on all scans with filtering at signal strength indices (SSIs) ≥ 7; training sets differed in size and SSI composition, and both models were tested on identical held-out samples stratified by SSI. Outcomes were pointwise mean absolute error (MAE) and absolute MD error. RESULTS: Mean SSI was 8.03 ± 1.65; 85.5% of scans had SSI ≥ 7. In the SSI < 7 subgroup, the all-SSI-trained model yielded lower errors for all endpoints, with reductions of 0.439 dB (24-2 MAE), 0.473 dB (10-2 MAE), 0.417 dB (24-2 MD error), and 0.453 dB (10-2 MD error; all adjusted P < 0.001). In the SSI ≥ 7 subgroup, no significant differences were observed. Error increased as SSI decreased, without a clear threshold at SSI = 7. CONCLUSIONS: In this EfficientNet 3D-based segmentation-free 3DCNN framework, using OCT scans across the full SSI range was associated with lower error in low-SSI held-out scans, with no significant penalty in higher SSI held-out scans in the primary comparison within this dataset. TRANSLATIONAL RELEVANCE: Including lower-signal-strength NIDEK macular OCT scans in development may improve robustness to lower-SSI inputs in this segmentation-free OCT-based VF framework.

PURPOSE: To investigate the natural course of geographic atrophy (GA) in Japanese patients, with a focus on factors associated with the change of GA size over time and visual acuity decline. STUDY DESIGN: Retrospective, single-center observational study. METHODS: We reviewed 40 eyes of 40 patients with GA and longitudinal fundus autofluorescence (FAF) imaging. Annual progression rates were calculated as square root-transformed (mm/year). Assessed factors included age, sex, smoking history, diabetes, hypertension, baseline best-corrected visual acuity (BCVA), baseline lesion size, foveal center involvement, choroidal thickness, FAF subtypes, and presence of subretinal drusenoid deposits (SDD). Visual outcomes were assessed as change in BCVA (ΔBCVA, logarithm of the minimum angle of resolution (logMAR)). RESULTS: Median age was 75.5 years; 50% of eyes had SDD. The median annual GA progression rate was 0.16 mm/year in square root units. Baseline GA area and the presence of SDD were independent predictors of GA progression (β = -0.0837, p = 0.015; β = +0.3064, p = 0.009, respectively). In multivariable analysis, baseline GA area also remained a significant predictor of visual deterioration (β = +0.1369, p = 0.002). When stratified by FAF patterns, there was significant worsening of BCVA in the Diffuse (p = 0.0027) and Banded (p = 0.0117) subtypes, whereas the None and Other subtypes did not show significant changes in BCVA over one year. CONCLUSIONS: SDD was identified as a significant risk factor for GA progression in Japanese patients. These findings improve understanding of natural history of GA in Japanese population and may inform risk stratification.

UNLABELLED: This study investigated the one-year real-world clinical outcomes of intravitreal faricimab (IVF) in treatment-naïve neovascular age-related macular degeneration (nAMD) and its association with aqueous humor cytokine profiles. Thirty-four eyes of 32 nAMD patients, who received initial IVF at Jichi Medical University were included. These eyes showed significant improvements in central retinal subfield thickness (CST), central choroidal thickness (CCT), and best-corrected visual acuity (BCVA) one year post-IVF (all P < 0.01). Among 24 measurable cytokines, only VEGF-A was significantly higher in eyes with a dry macula at week 16 (P = 0.0030). Three cytokines were significantly higher in eyes with longer injection interval. Higher levels of galectin-1 and VCAM-1, and lower levels of P-selectin, were correlated with greater CST reduction (P = 0.020, 0.0061, and 0.012, respectively). Conversely, five cytokines including Ang-1 were correlated with greater BCVA improvement (all P < 0.05). In conclusion, an anatomical perspective faricimab demonstrated anatomical efficacy for VEGF-driven nAMD in the loading phase, while multiple cytokines associated with vascular instability or fibrosis appear to contribute to its durability. Notably, Ang-1 may be linked to visual improvement, suggesting that the neurotrophic effects of Ang-1 could be enhanced by Ang-2 inhibition. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1038/s41598-025-28911-9.