Predicting Glaucoma Development With Longitudinal Deep Learning Predictions From Fundus Photographs.
Predicting Glaucoma Development With Longitudinal Deep Learning Predictions From Fundus Photographs.
复制标题
DOI：
10.1016/j.ajo.2020.12.031
复制
发表时间：
2021-05
影响因子：
4.2
通讯作者：
Medeiros FA
中科院分区：
文献类型：
--
作者：
Lee T;Jammal AA;Mariottoni EB;Medeiros FA
评估深度学习算法根据眼底照片预测视网膜神经纤维层 (RNFL) 厚度的纵向变化是否可以预测青光眼视野缺损的未来发展。回顾性队列研究 这项研究包括 827 名青光眼疑似患者的 1072 只眼睛，平均随访时间为 5.9 ± 3.8 年。所有眼睛的基线均具有正常的标准自动视野检查 (SAP)。在整个随访过程中获取了额外的 SAP 和眼底照片。转化为青光眼被定义为 SAP 上可重复的青光眼缺陷。使用 OCT 训练的深度学习算法（机器到机器，M2M）根据眼底照片预测 RNFL 厚度。联合纵向生存模型用于评估 M2M RNFL 厚度估计的基线和纵向变化是否可以预测视野缺损的发展。随访期间，196 只眼睛 (18%) 转化为青光眼。 M2M 预测的 RNFL 厚度的平均变化率为 -1.02 μm/y（对于转化者）和 -0.67 μm/y（对于非转化者）（P < 0.001）。基线和预测 RNFL 厚度的变化率可显着预测转化为青光眼，多变量模型中基线每降低 10 μm，风险比为 1.56（95% CI，1.33 – 1.82；P < 0.001），随访期间厚度每减少 1 μm/y，风险比为 1.99（95% CI，1.36 – 2.93；P < 0.001）。 0.001）。深度学习算法根据眼底照片预测 RNFL 厚度测量结果的纵向变化可用于预测疑似患有青光眼的眼睛发生青光眼的风险。
To assess whether longitudinal changes in a deep learning algorithm’s predictions of retinal nerve fiber layer (RNFL) thickness based on fundus photographs can predict future development of glaucomatous visual field defects. Retrospective cohort study This study included 1072 eyes of 827 glaucoma-suspect patients with an average follow-up of 5.9 ± 3.8 years. All eyes had normal standard automated perimetry (SAP) at baseline. Additional SAP and fundus photographs were acquired throughout follow-up. Conversion to glaucoma was defined as repeatable glaucomatous defects on SAP. An OCT-trained deep learning algorithm (Machine to Machine, M2M) was used to predict RNFL thicknesses from fundus photographs. Joint longitudinal survival models were used to assess whether baseline and longitudinal change in M2M’s RNFL thickness estimates could predict development of visual field defects. 196 eyes (18%) converted to glaucoma during follow-up. The mean rate of change in M2M’s predicted RNFL thickness was −1.02 μm/y for conversors and −0.67 μm/y for non-conversors (P < 0.001). Baseline and rate of change of predicted RNFL thickness were significantly predictive of conversion to glaucoma, with hazard ratios in the multivariable model of 1.56 per 10 μm lower at baseline (95% CI, 1.33 – 1.82; P < 0.001) and 1.99 per 1 μm/y faster loss in thickness during follow-up (95% CI, 1.36 – 2.93; P < 0.001). Longitudinal changes in a deep learning algorithm’s predictions of RNFL thickness measurements based on fundus photographs can be used to predict risk of glaucoma conversion in eyes suspected of having the disease.
登录
查看更多内容
影响因子：
4.7
作者：
Bursztyn, Lulu;Woodward, Maria A.;De Lott, Lindsey B.
通讯作者：
De Lott, Lindsey B.
DOI：
10.1093/biostatistics/1.4.465
发表时间：
2000-12-01
期刊：
Biostatistics (Oxford, England)
影响因子：
--
作者：
Henderson, R;Diggle, P;Dobson, A
通讯作者：
Dobson, A
影响因子：
--
作者：
SOMMER, A;TIELSCH, JM;SINGH, K
通讯作者：
SINGH, K
影响因子：
3.1
作者：
Srihatrai, Parinya;Hlowchitsieng, Thanita
通讯作者：
Hlowchitsieng, Thanita
光学相干断层扫描，用于初级保健办公室的高级筛查。
影响因子：
2.8
作者：
Shelton, Ryan L.;Jung, Woonggyu;Sayegh, Samir I.;McCormick, Daniel T.;Kim, Jeehyun;Boppart, Stephen A.
通讯作者：
Boppart, Stephen A.