Sano Kei, Nishijima Euido, Sumi Shunsuke, Noro Takahiko, Ogawa Shumpei, Igari Yuka, Iwase Aiko, Nakano Tadashi
Department of Ophthalmology, The Jikei University School of Medicine, Tokyo, Japan.
Institute for Quantitative Biosciences (IQB), University of Tokyo, Tokyo, Japan.
Ophthalmol Sci. 2025 Apr 28;5(6):100805. doi: 10.1016/j.xops.2025.100805. eCollection 2025 Nov-Dec.
To develop DeepISP, a deep learning model that predicts the comprehensive visual field (VF) information of the Humphrey visual field analyzer (HFA) based on rapid screening perimetry (Imo/TEMPO screening program [ISP]).
A retrospective, cross-sectional, and longitudinal cohort database study.
One hundred eighty-seven actual ISPs from 112 patients who underwent both ISP and HFA 24-2 on the same day at the Jikei University School of Medicine Affiliated Hospital and 3470 synthesized ISPs from 883 patients who underwent VF measurements using HFA 24-2 and HFA 10-2 at 4 hospitals affiliated with Jikei University School of Medicine.
We developed 2 variants of multitask neural networks designed to predict both current VF parameters and VF progression parameters. We also evaluated the efficacy of data augmentation to synthesize ISP tests created by combining 20 points from HFA 24-2 and 8 points from HFA 10-2, with thresholding applied to these 28 points.
Mean absolute error for mean deviation (MD), pattern standard deviation (PSD), and visual field index (VFI). Mean F1 score for total deviation (TD) and pattern deviation (PD) probability plot classification. Area under the curve (AUC) for MD progression (MD slope <-1.0 decibel/year) and VFI progression (VFI slope <-1.8%/year).
DeepISP could predict current VF status. Mean absolute errors for predicting MD, PSD, and VFI were 1.869 ± 0.114, 1.918 ± 0.082, and 5.146 ± 0.487, respectively. The mean F1 scores for pointwise classification of TD and PD probability plots were 0.761 ± 0.002 and 0.775 ± 0.002, respectively. The AUC for classifying glaucoma hemifield test was 0.920 ± 0.008. DeepISP was also capable of predicting VF progression, with AUCs of 0.828 ± 0.060 and 0.832 ± 0.062 for predicting MD and VFI progression, respectively.
We demonstrated ISP's versatility and capability in predicting comprehensive VF information, including current severity and progression risk. Our DeepISP serves as an efficient tool for screening and prioritizing patients with glaucoma for clinical intervention using only a single rapid ISP test.
Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
开发DeepISP，一种深度学习模型，可基于快速筛查视野计（Imo/TEMPO筛查程序[ISP]）预测 Humphrey视野分析仪（HFA）的综合视野（VF）信息。
一项回顾性、横断面和纵向队列数据库研究。
来自112名患者的187次实际ISP，这些患者同一天在东京慈惠会医科大学附属医院接受了ISP和HFA 24-2检查；以及来自883名患者的3470次合成ISP，这些患者在东京慈惠会医科大学附属的4家医院接受了HFA 24-2和HFA 10-2视野测量。
我们开发了2种多任务神经网络变体，旨在预测当前VF参数和VF进展参数。我们还评估了数据增强的效果，以合成通过将HFA 24-2的20个点和HFA 10-2的8个点组合而成的ISP测试，并对这28个点应用阈值处理。
平均偏差（MD）、模式标准差（PSD）和视野指数（VFI）的平均绝对误差。总偏差（TD）和模式偏差（PD）概率图分类的平均F1分数。MD进展（MD斜率<-1.0分贝/年）和VFI进展（VFI斜率<-1.8%/年）的曲线下面积（AUC）。
DeepISP可以预测当前VF状态。预测MD、PSD和VFI的平均绝对误差分别为1.869±0.114、1.918±0.082和5.146±0.487。TD和PD概率图逐点分类的平均F1分数分别为0.761±0.002和0.775±0.002。青光眼半视野检查分类的AUC为0.920±0.008。DeepISP还能够预测VF进展，预测MD和VFI进展的AUC分别为0.828±0.060和0.832±0.062。
我们证明了ISP在预测综合VF信息（包括当前严重程度和进展风险）方面的多功能性和能力。我们的DeepISP是一种高效工具，仅通过一次快速ISP测试即可对青光眼患者进行筛查并确定临床干预的优先级。
本文末尾的脚注和披露中可能会有专有或商业披露信息。
