生成逼真眼底照片以改进自动疾病分类
本研究训练去噪扩散概率模型(DDPM)生成代表糖尿病视网膜病变、年龄相关性黄斑变性、青光眼或正常眼的视网膜眼底图像,由8名委员会认证眼科医生评估96张测试图像的真实性,并用100至1000张生成图像扩充训练深度卷积神经网络集成,发现眼科医生区分真实与生成图像的平均准确率为61.1%,在最小类别中加入238张生成图像使F-score和准确率分别提升5.3%和5.8%(p<0.01)。
要点
- DDPM可生成高度逼真的视网膜眼底图像,眼科医生区分真实与生成图像的平均准确率仅61.1%。
- 在最小类别中加入238张生成图像,使视网膜疾病分类的F-score和准确率分别提升5.3%和5.8%(p<0.01)。
- 生成图像扩充训练集可提升CNN集成性能,无需额外真实患者数据。
核心结论
- 潜扩散模型生成的视网膜图像高度逼真,经人类专家验证。
- 将生成图像加入训练集可提升CNN集成性能,且无需额外真实患者数据。
研究设计与样本
- 训练DDPM生成代表糖尿病视网膜病变、年龄相关性黄斑变性、青光眼或正常眼的视网膜眼底图像。
- 8名委员会认证眼科医生评估96张测试图像的真实性并按疾病标签分类。
- 使用100至1000张生成图像扩充深度卷积集成训练以分类视网膜疾病。
- 测量眼科医生正确识别真实与生成图像的准确率。
- 测量训练后CNN在100张眼底图像测试集上分类视网膜疾病的准确率、F-score和受试者工作特征曲线下面积。
主要发现
- 眼科医生区分真实与生成图像的平均准确率为61.1%(范围:51.0%-68.8%)。
- 在最小类别中加入238张生成图像后,视网膜疾病分类任务的F-score和准确率分别 statistically significantly 提升5.3%和5.8%(p<0.01),与仅用真实图像训练的基线模型相比。
关键数值与推荐
- 眼科医生区分真实与生成图像的平均准确率为61.1%(范围:51.0%-68.8%)。
- 加入238张生成图像使F-score提升5.3%、准确率提升5.8%(p<0.01)。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
AIMS: This study aims to investigate whether denoising diffusion probabilistic models (DDPMs) could generate realistic retinal images, and if they could be used to improve the performance of a deep convolutional neural network (CNN) ensemble for multiple retinal disease classification, which was previously shown to outperform human experts. METHODS: We trained DDPMs to generate retinal fundus images representing diabetic retinopathy, age-related macular degeneration, glaucoma or normal eyes. Eight board-certified ophthalmologists evaluated 96 test images to assess the realism of generated images and classified them based on disease labels. Subsequently, between 100 and 1000 generated images were employed to augment training of deep convolutional ensembles for classifying retinal disease. We measured the accuracy of ophthalmologists in correctly identifying real and generated images. We also measured the classification accuracy, F-score and area under the receiver operating curve of a trained CNN in classifying retinal diseases from a test set of 100 fundus images. RESULTS: Ophthalmologists exhibited a mean accuracy of 61.1% (range: 51.0%-68.8%) in differentiating real and generated images. Augmenting the training set with 238 generated images in the smallest class statistically significantly improved the F-score and accuracy by 5.3% and 5.8%, respectively (p<0.01) in a retinal disease classification task, compared with a baseline model trained only with real images. CONCLUSIONS: Latent diffusion models generated highly realistic retinal images, as validated by human experts. Adding generated images to the training set improved performance of a CNN ensemble without requiring additional real patient data.
引用
- Pandey PU, Micieli JA, Ong Tone S, et al. Realistic fundus photograph generation for improving automated disease classification. Br J Ophthalmol. 2025. PMID: 39939121. doi:10.1136/bjo-2024-326122
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。