UNASSIGNED:开发和验证基于深度学习(DL)的自动化人工智能(AI)平台,用于使用基于镜片不透明度分类系统(LOCS)III的裂隙灯和逆透镜照片对白内障进行诊断和分级。
UNASSIGNED:横断面研究,其中使用裂隙灯照片和逆透镜照片对卷积神经网络进行训练和测试。
UNASISIGNED:来自596名患者的一千三百三十五个裂隙灯图像和637个逆向照明透镜图像。
UNASSIGNED:由2名经过培训的分级人员使用LOCSIII对裂隙灯和后照镜头照片进行分级。图像数据集被标记并划分为训练,验证,和测试数据集。我们在AI领域使用4个关键策略对AI平台进行了训练和验证:(1)数据内部冗余信息的区域检测网络,(2)小数据集问题的数据扩充和迁移学习,(3)数据集偏差的广义交叉熵损失,(4)类不平衡问题的类平衡损失。AI平台的性能得到了3种AI算法的整合:ResNet18,WideResNet50-2和ResNext50。
UNASSIGNED:基于诊断和LOCSIII的AI平台分级预测性能。
UNASSIGNED:AI平台显示出强大的诊断性能(接收器工作特性曲线下的面积[AUC],0.9992[95%置信区间(CI),0.9986-0.9998]和0.9994[95%CI,0.9989-0.9998];准确性,98.82%[95%CI,97.7%-99.9%]和98.51%[95%CI,97.4%-99.6%])和基于LOCSIII的分级预测性能(AUC,0.9567[95%CI,0.9501-0.9633]和0.9650[95%CI,0.9509-0.9792];准确性,91.22%[95%CI,89.4%-93.0%]和90.26%[95%CI,88.6%-91.9%])使用裂隙灯照片进行核乳光(NO)和核颜色(NC),分别。对于皮质混浊(CO)和后囊膜下混浊(PSC),系统实现了高诊断性能(AUC,0.9680[95%CI,0.9579-0.9781]和0.9465[95%CI,0.9348-0.9582];准确性,96.21%[95%CI,94.4%-98.0%]和92.17%[95%CI,88.6%-95.8%])和良好的基于LOCSIII的分级预测性能(AUC,0.9044[95%CI,0.8958-0.9129]和0.9174[95%CI,0.9055-0.9295];准确性,91.33%[95%CI,89.7%-93.0%]和87.89%[95%CI,85.6%-90.2%])使用逆向照明图像。
UNASSIGNED:我们基于DL的AI平台成功地在7级分类中对NO和NC以及6级分类中对CO和PSC进行了准确和精确的检测和分级,克服了医学数据库的局限性,例如训练数据很少或标签分布有偏差。
UNASSIGNED: To develop and validate an automated deep learning (DL)-based artificial intelligence (AI) platform for diagnosing and grading cataracts using slit-lamp and retroillumination lens photographs based on the Lens Opacities Classification System (LOCS) III.
UNASSIGNED: Cross-sectional study in which a convolutional neural network was trained and tested using photographs of slit-lamp and retroillumination lens photographs.
UNASSIGNED: One thousand three hundred thirty-five slit-lamp images and 637 retroillumination lens images from 596 patients.
UNASSIGNED: Slit-lamp and retroillumination lens photographs were graded by 2 trained graders using LOCS III. Image datasets were labeled and divided into training, validation, and test datasets. We trained and validated AI platforms with 4 key strategies in the AI domain: (1) region detection network for redundant information inside data, (2) data augmentation and transfer learning for the small dataset size problem, (3) generalized cross-entropy loss for dataset bias, and (4) class balanced loss for class imbalance problems. The performance of the AI platform was reinforced with an ensemble of 3 AI algorithms: ResNet18, WideResNet50-2, and ResNext50.
UNASSIGNED: Diagnostic and LOCS III-based grading prediction performance of AI platforms.
UNASSIGNED: The AI platform showed robust diagnostic performance (area under the receiver operating characteristic curve [AUC], 0.9992 [95% confidence interval (CI), 0.9986-0.9998] and 0.9994 [95% CI, 0.9989-0.9998]; accuracy, 98.82% [95% CI, 97.7%-99.9%] and 98.51% [95% CI, 97.4%-99.6%]) and LOCS III-based grading prediction performance (AUC, 0.9567 [95% CI, 0.9501-0.9633] and 0.9650 [95% CI, 0.9509-0.9792]; accuracy, 91.22% [95% CI, 89.4%-93.0%] and 90.26% [95% CI, 88.6%-91.9%]) for nuclear opalescence (NO) and nuclear color (NC) using slit-lamp photographs, respectively. For cortical opacity (CO) and posterior subcapsular opacity (PSC), the system achieved high diagnostic performance (AUC, 0.9680 [95% CI, 0.9579-0.9781] and 0.9465 [95% CI, 0.9348-0.9582]; accuracy, 96.21% [95% CI, 94.4%-98.0%] and 92.17% [95% CI, 88.6%-95.8%]) and good LOCS III-based grading prediction performance (AUC, 0.9044 [95% CI, 0.8958-0.9129] and 0.9174 [95% CI, 0.9055-0.9295]; accuracy, 91.33% [95% CI, 89.7%-93.0%] and 87.89% [95% CI, 85.6%-90.2%]) using retroillumination images.
UNASSIGNED: Our DL-based AI platform successfully yielded accurate and precise detection and grading of NO and NC in 7-level classification and CO and PSC in 6-level classification, overcoming the limitations of medical databases such as few training data or biased label distribution.