PDF(Pubmed)
Abstract：
UNASSIGNED: A deep learning model was developed to detect nonexudative macular neovascularization (neMNV) using OCT B-scans.
UNASSIGNED: Retrospective review of a prospective, observational study.
UNASSIGNED: Normal control eyes and patients with age-related macular degeneration (AMD) with and without neMNV.
UNASSIGNED: Swept-source OCT angiography (SS-OCTA) imaging (PLEX Elite 9000, Carl Zeiss Meditec, Inc) was performed using the 6 × 6-mm scan pattern. Individual B-scans were annotated to distinguish between drusen and the double-layer sign (DLS) associated with the neMNV. The machine learning model was tested on a dataset graded by humans, and model performance was compared with the human graders.
UNASSIGNED: Intersection over Union (IoU) score was measured to evaluate segmentation network performance. Area under the receiver operating characteristic curve values, sensitivity, specificity, and positive predictive value (PPV) and negative predictive value (NPV) were measured to assess the performance of the final classification performance. Chance-corrected agreement between the algorithm and the human grader determinations was measured with Cohen\'s kappa.
UNASSIGNED: A total of 251 eyes from 210 patients, including 182 eyes with DLS and 115 eyes with drusen, were used for model training. Of 125 500 B-scans, 6879 B-scans were manually annotated. A vision transformer segmentation model was built to extract DLS and drusen from B-scans. The extracted prediction masks from all B-scans in a volume were projected to an en face image, and an eye-level projection map was obtained for each eye. A binary classification algorithm was established to identify eyes with neMNV from the projection map. The algorithm achieved 82%, 90%, 79%, and 91% sensitivity, specificity, PPV, and NPV, respectively, on a separate test set of 100 eyes that were evaluated by human graders in a previous study. The area under the curve value was calculated as 0.91 (95% confidence interval, 0.85-0.98). The results of the algorithm showed excellent agreement with the senior human grader (kappa = 0.83, P < 0.001) and moderate agreement with the junior grader consensus (kappa = 0.54, P < 0.001).
UNASSIGNED: Our network (code is available at https://github.com/uw-biomedical-ml/double_layer_vit) was able to detect the presence of neMNV from structural B-scans alone by applying a purely transformer-based model.
UNASSIGNED: Retrospective review of a prospective, observational study.
UNASSIGNED: Normal control eyes and patients with age-related macular degeneration (AMD) with and without neMNV.
UNASSIGNED: Swept-source OCT angiography (SS-OCTA) imaging (PLEX Elite 9000, Carl Zeiss Meditec, Inc) was performed using the 6 × 6-mm scan pattern. Individual B-scans were annotated to distinguish between drusen and the double-layer sign (DLS) associated with the neMNV. The machine learning model was tested on a dataset graded by humans, and model performance was compared with the human graders.
UNASSIGNED: Intersection over Union (IoU) score was measured to evaluate segmentation network performance. Area under the receiver operating characteristic curve values, sensitivity, specificity, and positive predictive value (PPV) and negative predictive value (NPV) were measured to assess the performance of the final classification performance. Chance-corrected agreement between the algorithm and the human grader determinations was measured with Cohen\'s kappa.
UNASSIGNED: A total of 251 eyes from 210 patients, including 182 eyes with DLS and 115 eyes with drusen, were used for model training. Of 125 500 B-scans, 6879 B-scans were manually annotated. A vision transformer segmentation model was built to extract DLS and drusen from B-scans. The extracted prediction masks from all B-scans in a volume were projected to an en face image, and an eye-level projection map was obtained for each eye. A binary classification algorithm was established to identify eyes with neMNV from the projection map. The algorithm achieved 82%, 90%, 79%, and 91% sensitivity, specificity, PPV, and NPV, respectively, on a separate test set of 100 eyes that were evaluated by human graders in a previous study. The area under the curve value was calculated as 0.91 (95% confidence interval, 0.85-0.98). The results of the algorithm showed excellent agreement with the senior human grader (kappa = 0.83, P < 0.001) and moderate agreement with the junior grader consensus (kappa = 0.54, P < 0.001).
UNASSIGNED: Our network (code is available at https://github.com/uw-biomedical-ml/double_layer_vit) was able to detect the presence of neMNV from structural B-scans alone by applying a purely transformer-based model.
摘要：
UNASSIGNED：开发了一种深度学习模型，用于使用OCTB扫描检测非渗出性黄斑新生血管(neMNV)。
未经评估：前瞻性回顾，观察性研究。
UNASSIGNED：正常对照眼睛和患有和不患有neMNV的年龄相关性黄斑变性（AMD）的患者。
UNASSIGNED：扫描源OCT血管造影（SS-OCTA）成像（PLEXElite9000，CarlZeissMeditec，Inc）使用6×6-mm扫描图案进行。对单个B扫描进行注释以区分玻璃疣和与neMNV相关的双层标志（DLS）。机器学习模型是在由人类分级的数据集上测试的，并将模型性能与人类分级者进行了比较。
UNASSIGNED：测量联合交集(IoU)评分以评估分段网络性能。接收器工作特性曲线值下的面积,灵敏度,特异性，测量阳性预测值(PPV)和阴性预测值(NPV)以评估最终分类性能。使用Cohen的kappa测量算法与人类分级者确定之间的机会校正一致性。
未经证实：共有210名患者的251只眼，包括182只DLS的眼睛和115只玻璃疣的眼睛，用于模型训练。125500次B扫描，手动注释6879个B扫描。建立了视觉变压器分割模型，从B扫描中提取DLS和玻璃疣。从体积中的所有B扫描中提取的预测掩模被投影到en面部图像，并获得每只眼睛的眼睛水平投影图。建立了二元分类算法，从投影图中识别具有neMNV的眼睛。该算法取得了82%,90%,79%,和91%的灵敏度，特异性，PPV,和净现值，分别,在先前研究中由人类分级者评估的100只眼睛的单独测试集上。曲线下面积值计算为0.91(95%置信区间，0.85-0.98）。该算法的结果显示与高级人类等级者的良好一致性（kappa=0.83，P<0.001），与初级等级者的一致性中等（kappa=0.54，P<0.001）。
UNASSIGNED：我们的网络（代码可在https://github.com/uw-biomedical-ml/double_layer_vit上获得）通过应用纯基于变压器的模型，能够从结构B扫描中检测到neMNV的存在。
未经评估：前瞻性回顾，观察性研究。
UNASSIGNED：正常对照眼睛和患有和不患有neMNV的年龄相关性黄斑变性（AMD）的患者。
UNASSIGNED：扫描源OCT血管造影（SS-OCTA）成像（PLEXElite9000，CarlZeissMeditec，Inc）使用6×6-mm扫描图案进行。对单个B扫描进行注释以区分玻璃疣和与neMNV相关的双层标志（DLS）。机器学习模型是在由人类分级的数据集上测试的，并将模型性能与人类分级者进行了比较。
UNASSIGNED：测量联合交集(IoU)评分以评估分段网络性能。接收器工作特性曲线值下的面积,灵敏度,特异性，测量阳性预测值(PPV)和阴性预测值(NPV)以评估最终分类性能。使用Cohen的kappa测量算法与人类分级者确定之间的机会校正一致性。
未经证实：共有210名患者的251只眼，包括182只DLS的眼睛和115只玻璃疣的眼睛，用于模型训练。125500次B扫描，手动注释6879个B扫描。建立了视觉变压器分割模型，从B扫描中提取DLS和玻璃疣。从体积中的所有B扫描中提取的预测掩模被投影到en面部图像，并获得每只眼睛的眼睛水平投影图。建立了二元分类算法，从投影图中识别具有neMNV的眼睛。该算法取得了82%,90%,79%,和91%的灵敏度，特异性，PPV,和净现值，分别,在先前研究中由人类分级者评估的100只眼睛的单独测试集上。曲线下面积值计算为0.91(95%置信区间，0.85-0.98）。该算法的结果显示与高级人类等级者的良好一致性（kappa=0.83，P<0.001），与初级等级者的一致性中等（kappa=0.54，P<0.001）。
UNASSIGNED：我们的网络（代码可在https://github.com/uw-biomedical-ml/double_layer_vit上获得）通过应用纯基于变压器的模型，能够从结构B扫描中检测到neMNV的存在。
