Faculty of Mechanical EngineeringTechnion Israel Institute of Technology Haifa 3200003 Israel.
Department of OphthalmologyRambam Health Care Campus Haifa 3109601 Israel.
IEEE J Transl Eng Health Med. 2023 Jul 24;11:487-494. doi: 10.1109/JTEHM.2023.3294904. eCollection 2023.
Prospective evaluation of OCT images of DME (n = 320) subject to elastic transformation, with the deformation intensity represented by ([Formula: see text]). Three sets of images, each comprising 100 pairs of scans (100 original & 100 modified), were grouped according to the range of ([Formula: see text]), including low-, medium- and high-degree of augmentation; ([Formula: see text] = 1-6), ([Formula: see text] = 7-12), and ([Formula: see text] = 13-18), respectively. Three retina specialists evaluated all datasets in a blinded manner and designated each image as 'original' versus 'modified'. The rate of assignment of 'original' value to modified images (false-negative) was determined for each grader in each dataset.
The false-negative rates ranged between 71-77% for the low-, 63-76% for the medium-, and 50-75% for the high-augmentation categories. The corresponding rates of correct identification of original images ranged between 75-85% ([Formula: see text]0.05) in the low-, 73-85% ([Formula: see text]0.05 for graders 1 & 2, p = 0.01 for grader 3) in the medium-, and 81-91% ([Formula: see text]) in the high-augmentation categories. In the subcategory ([Formula: see text] = 7-9) the false-negative rates were 93-83%, whereas the rates of correctly identifying original images ranged between 89-99% ([Formula: see text]0.05 for all graders).
Deformation of low-medium intensity ([Formula: see text] = 1-9) may be applied without compromising OCT image representativeness in DME. Clinical and Translational Impact Statement-Elastic deformation may efficiently augment the size, robustness, and diversity of training datasets without altering their clinical value, enhancing the development of high-accuracy algorithms for automated interpretation of OCT images.
探索光学相干断层扫描（OCT）图像弹性变形的临床有效性，以增加数据量，为基于深度学习的糖尿病黄斑水肿（DME）检测模型的开发提供支持。
对 320 例 DME 的 OCT 图像进行前瞻性评估，使用变形强度[Formula: see text]表示。将图像分为三组，每组包含 100 对扫描（100 个原始图像和 100 个变形图像），根据变形强度范围[Formula: see text]分为低、中、高强度变形，分别为[Formula: see text] = 1-6，[Formula: see text] = 7-12 和 [Formula: see text] = 13-18。三位视网膜专家以盲法对所有数据集进行评估，并将每张图像指定为“原始”或“变形”。在每个数据集和每个评分者中，确定将“原始”值分配给变形图像（假阴性）的比例。
低强度变形组的假阴性率为 71-77%，中强度变形组为 63-76%，高强度变形组为 50-75%。正确识别原始图像的比例分别为低强度变形组 75-85%（[Formula: see text]0.05），中强度变形组 73-85%（[Formula: see text]0.05，评分者 1 和 2；评分者 3 为 p = 0.01），高强度变形组 81-91%（[Formula: see text]）。在[Formula: see text] = 7-9 亚组中，假阴性率为 93-83%，而正确识别原始图像的比例分别为 89-99%（所有评分者均为[Formula: see text]0.05）。
低-中强度变形（[Formula: see text] = 1-9）可应用于 OCT 图像，不会改变 DME 图像的代表性。
弹性变形可有效增加训练数据集的规模、稳健性和多样性，同时不改变其临床价值，从而促进用于自动解释 OCT 图像的高精度算法的开发。
