Department of Ophthalmology, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan.
Department of Ophthalmology, Seirei Hamamatsu General Hospital, Hamamatsu, Shizuoka, Japan.
Transl Vis Sci Technol. 2024 Oct 1;13(10):35. doi: 10.1167/tvst.13.10.35.
Our previous study predicted genuine glaucomatous visual field (VF) impairment in the central 10° VF, excluding the effect of cataract, using visual acuity (VA) and global indexes of VF more accurately than pattern deviation (PD). This study aimed to improve the accuracy by using pointwise total deviation (TD) values with the machine-learning method of random forest model (RFM) and to investigate whether incorporating optical coherence tomography-measured ganglion cell-inner plexiform layer (GCIPL) thickness is useful.
This retrospective study included 89 eyes with open-angle glaucoma that underwent successful cataract surgery (with or without iStent implantation or ab interno trabeculotomy). Postoperative TD in each of the 68 VF points was predicted using preoperative (1) PD, (2) VA and VF with a linear regression model (LM), and (3) VA and VF with RFM, and averaged as predicted mean TD (mTDpost). Further prediction was made by incorporating the preoperative GCIPL into the best model.
The mean absolute error (MAE) between the actual and predicted mTDpost with RFM (1.25 ± 1.03 dB) was significantly smaller than that with PD (3.20 ± 4.06 dB, p < 0.01) and LM (1.42 ± 1.06 dB, p < 0.05). The MAEs with the model incorporating GCIPL into RFM (1.24 ± 1.04 dB) and RFM were not significantly different.
Accurate prediction of genuine glaucomatous VF impairment was achieved using pointwise TD with RFM. No merit was observed by incorporating the GCIPL into this model.
This pointwise RFM could clinically reduce cataract effect on VF.
我们之前的研究表明，使用视力 (VA) 和视场 (VF) 的全局指标比图形偏差 (PD) 更准确地预测中央 10° VF 中的真正青光眼视野 (VF) 损害，且不考虑白内障的影响。本研究旨在通过使用机器学习方法随机森林模型 (RFM) 的逐点总偏差 (TD) 值来提高准确性，并研究是否纳入光学相干断层扫描测量的神经节细胞-内丛状层 (GCIPL) 厚度有用。
本回顾性研究纳入了 89 只接受过成功白内障手术（伴或不伴 iStent 植入或内路小梁切开术）的开角型青光眼眼。使用术前（1）PD、（2）VA 和 VF 的线性回归模型 (LM) 以及（3）VA 和 VF 的随机森林模型 (RFM) 预测术后每个 68 个 VF 点的 TD，平均作为预测平均 TD (mTDpost)。通过将术前 GCIPL 纳入最佳模型进一步进行预测。
RFM 预测的实际与预测 mTDpost 之间的平均绝对误差 (MAE) (1.25 ± 1.03 dB) 明显小于 PD (3.20 ± 4.06 dB，p < 0.01) 和 LM (1.42 ± 1.06 dB，p < 0.05)。将 GCIPL 纳入 RFM 模型的 MAE（1.24 ± 1.04 dB）与 RFM 模型的 MAE 无显著差异。
使用 RFM 进行逐点 TD 可准确预测真正的青光眼 VF 损害。将 GCIPL 纳入该模型没有观察到优势。
医脉通
