DeepLensNet:深度学习自动诊断与白内障类型及严重程度的定量分类
DeepLensNet 基于年龄相关性眼病研究(AREDS)数据集训练深度学习模型,从眼前节照片自动诊断和定量分类年龄相关性白内障。在测试集富集阳性病例的子集中,核硬化(NS)和皮质晶状体混浊(CLO)的平均均方误差(MSE)显著优于眼科医生和医学生;后囊下白内障(PSC)与眼科医生相似。外部验证使用新加坡马来眼研究数据。
要点
- DeepLensNet 可从眼前节照片自动定量分类核硬化、皮质晶状体混浊和后囊下白内障。
- 在富集阳性病例的子集中,NS 和 CLO 的 MSE 显著低于眼科医生和医学生。
- PSC 的 MSE 与眼科医生相似,与医学生无显著差异。
核心结论
- DeepLensNet 能对全部3种年龄相关性白内障进行自动定量严重程度分类。
- 对最常见的2种类型(NS和CLO),其准确性显著优于眼科医生;对最不常见的类型(PSC),与眼科医生相似。
- DeepLensNet 可能在临床和研究领域有广泛潜在应用,未来此类方法可能提高全球白内障评估的可及性。
研究设计与样本
- DeepLensNet 通过将深度学习模型应用于年龄相关性眼病研究(AREDS)数据集进行训练。
- 共纳入来自576名AREDS参与者1137只眼纵向随访的18999张照片(6333组三联照片)。
- 模型从45度裂隙灯照片检测和量化核硬化(NS;量表0.9-7.1),从后照法照片检测和量化皮质晶状体混浊(CLO;量表0%-100%)和后囊下白内障(PSC;量表0%-100%)。
- DeepLensNet 性能与14名眼科医生和24名医学生进行比较。
- 主要结局指标为均方误差(MSE)。
主要发现
- 在完整测试集上,DeepLensNet 的平均 MSE 为 NS 0.23(标准差[SD] 0.01)、CLO 13.1(SD 1.6)、PSC 16.6(SD 2.4)。
- 在测试集子集(大幅富集CLO和PSC阳性病例)中,NS 的平均 MSE 为 DeepLensNet 0.23(SD 0.02) vs 眼科医生 0.98(SD 0.24;P=0.000001) vs 医学生 1.24(SD 0.34;P=0.000005)。
- 在该子集中,CLO 的平均 MSE 为 DeepLensNet 53.5(SD 14.8) vs 眼科医生 134.9(SD 89.9;P=0.003) vs 医学生 433.6(SD 962.1;P=0.0007)。
- 在该子集中,PSC 的平均 MSE 为 DeepLensNet 171.9(SD 38.9) vs 眼科医生 176.8(SD 98.0;P=0.67) vs 医学生 398.2(SD 645.4;P=0.18)。
- 在新加坡马来眼研究的外部验证中(采样以反映AREDS的白内障严重程度分布),DeepSeeNet 的 MSE 为 NS 1.27、PSC 25.5。
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
PURPOSE: To develop deep learning models to perform automated diagnosis and quantitative classification of age-related cataract from anterior segment photographs. DESIGN: DeepLensNet was trained by applying deep learning models to the Age-Related Eye Disease Study (AREDS) dataset. PARTICIPANTS: A total of 18 999 photographs (6333 triplets) from longitudinal follow-up of 1137 eyes (576 AREDS participants). METHODS: Deep learning models were trained to detect and quantify nuclear sclerosis (NS; scale 0.9-7.1) from 45-degree slit-lamp photographs and cortical lens opacity (CLO; scale 0%-100%) and posterior subcapsular cataract (PSC; scale 0%-100%) from retroillumination photographs. DeepLensNet performance was compared with that of 14 ophthalmologists and 24 medical students. MAIN OUTCOME MEASURES: Mean squared error (MSE). RESULTS: On the full test set, mean MSE for DeepLensNet was 0.23 (standard deviation [SD], 0.01) for NS, 13.1 (SD, 1.6) for CLO, and 16.6 (SD, 2.4) for PSC. On a subset of the test set (substantially enriched for positive cases of CLO and PSC), for NS, mean MSE for DeepLensNet was 0.23 (SD, 0.02), compared with 0.98 (SD, 0.24; P = 0.000001) for the ophthalmologists and 1.24 (SD, 0.34; P = 0.000005) for the medical students. For CLO, mean MSE was 53.5 (SD, 14.8), compared with 134.9 (SD, 89.9; P = 0.003) for the ophthalmologists and 433.6 (SD, 962.1; P = 0.0007) for the medical students. For PSC, mean MSE was 171.9 (SD, 38.9), compared with 176.8 (SD, 98.0; P = 0.67) for the ophthalmologists and 398.2 (SD, 645.4; P = 0.18) for the medical students. In external validation on the Singapore Malay Eye Study (sampled to reflect the cataract severity distribution in AREDS), the MSE for DeepSeeNet was 1.27 for NS and 25.5 for PSC. CONCLUSIONS: DeepLensNet performed automated and quantitative classification of cataract severity for all 3 types of age-related cataract. For the 2 most common types (NS and CLO), the accuracy was significantly superior to that of ophthalmologists; for the least common type (PSC), it was similar. DeepLensNet may have wide potential applications in both clinical and research domains. In the future, such approaches may increase the accessibility of cataract assessment globally. The code and models are available at https://github.com/ncbi/deeplensnet.
引用
- Keenan TDL, Chen Q, Agrón E, et al. DeepLensNet: Deep Learning Automated Diagnosis and Quantitative Classification of Cataract Type and Severity. Ophthalmology. 2022. PMID: 34990643. doi:10.1016/j.ophtha.2021.12.017
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。