文献速递角膜与眼表疾病白内障与屈光手术青光眼

基于智能手机的深度学习模型用于白内障及多种角膜疾病的广泛诊断与分诊

Deep learning model for extensive smartphone-based diagnosis and triage of cataracts and multiple corneal diseases.
PubMed 文献 · Br J Ophthalmol · 2024-09-20

本研究利用裂隙灯显微镜图像(6106张)和智能手机图像(336张)共6442张图像,开发了一种人工智能算法,将36种主要疾病(白内障和角膜疾病)分为9个类别。使用智能手机图像测试显示,该算法在各疾病诊断中ROC曲线下面积达0.960至1.000,在转诊建议分诊中表现出高敏感度和特异度,结论认为该AI系统在白内障和角膜疾病诊断中展现出有前景的性能。

要点

  • AI算法基于裂隙灯图像开发,可区分36种主要疾病(白内障和角膜疾病)为9类
  • 智能手机图像测试中,各疾病诊断的ROC曲线下面积为0.960至1.000
  • 分诊转诊建议中,紧急、半紧急、常规和观察的敏感度和特异度均较高

核心结论

  • 该AI系统在白内障和角膜疾病的诊断中展现出有前景的性能。

研究设计与样本

  • 本研究纳入6442张图像,包括裂隙灯显微镜图像6106张和智能手机图像336张。
  • 基于裂隙灯图像开发AI算法,将36种主要疾病(白内障和角膜疾病)分为9个类别。
  • 使用智能手机图像作为测试数据集验证AI模型,评估敏感度、特异度和ROC曲线。

主要发现

  • AI算法在正常眼、感染性角膜炎、免疫性角膜炎、角膜瘢痕、眼表肿瘤、角膜沉积物、急性闭角型青光眼、白内障和大疱性角膜病变的ROC曲线下面积分别为0.998(95%CI 0.992至0.999)、0.986(95%CI 0.978至0.997)、0.960(95%CI 0.925至0.994)、0.987(95%CI 0.978至0.996)、0.997(95%CI 0.992至1.000)、0.993(95%CI 0.984至1.000)、1.000(95%CI 1.000至1.000)、0.992(95%CI 0.985至0.999)和0.993(95%CI 0.985至1.000)。
  • 智能手机图像分诊转诊建议中,紧急的敏感度和特异度分别为1.00(95%CI 0.478至1.00)和1.00(95%CI 0.976至1.000),半紧急分别为0.867(95%CI 0.683至0.962)和1.00(95%CI 0.971至1.000),常规分别为0.853(95%CI 0.689至0.950)和0.983(95%CI 0.942至0.998),观察分别为1.00(95%CI 0.958至1.00)和0.896(95%CI 0.797至0.957)。

关键数值与推荐

  • AI算法诊断正常眼的ROC曲线下面积为0.998(95%CI 0.992至0.999)。
  • AI算法诊断感染性角膜炎的ROC曲线下面积为0.986(95%CI 0.978至0.997)。
  • AI算法诊断免疫性角膜炎的ROC曲线下面积为0.960(95%CI 0.925至0.994)。
  • AI算法诊断角膜瘢痕的ROC曲线下面积为0.987(95%CI 0.978至0.996)。
  • AI算法诊断眼表肿瘤的ROC曲线下面积为0.997(95%CI 0.992至1.000)。

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

AIM: To develop an artificial intelligence (AI) algorithm that diagnoses cataracts/corneal diseases from multiple conditions using smartphone images. METHODS: This study included 6442 images that were captured using a slit-lamp microscope (6106 images) and smartphone (336 images). An AI algorithm was developed based on slit-lamp images to differentiate 36 major diseases (cataracts and corneal diseases) into 9 categories. To validate the AI model, smartphone images were used for the testing dataset. We evaluated AI performance that included sensitivity, specificity and receiver operating characteristic (ROC) curve for the diagnosis and triage of the diseases. RESULTS: The AI algorithm achieved an area under the ROC curve of 0.998 (95% CI, 0.992 to 0.999) for normal eyes, 0.986 (95% CI, 0.978 to 0.997) for infectious keratitis, 0.960 (95% CI, 0.925 to 0.994) for immunological keratitis, 0.987 (95% CI, 0.978 to 0.996) for cornea scars, 0.997 (95% CI, 0.992 to 1.000) for ocular surface tumours, 0.993 (95% CI, 0.984 to 1.000) for corneal deposits, 1.000 (95% CI, 1.000 to 1.000) for acute angle-closure glaucoma, 0.992 (95% CI, 0.985 to 0.999) for cataracts and 0.993 (95% CI, 0.985 to 1.000) for bullous keratopathy. The triage of referral suggestion using the smartphone images exhibited high performance, in which the sensitivity and specificity were 1.00 (95% CI, 0.478 to 1.00) and 1.00 (95% CI, 0.976 to 1.000) for 'urgent', 0.867 (95% CI, 0.683 to 0.962) and 1.00 (95% CI, 0.971 to 1.000) for 'semi-urgent', 0.853 (95% CI, 0.689 to 0.950) and 0.983 (95% CI, 0.942 to 0.998) for 'routine' and 1.00 (95% CI, 0.958 to 1.00) and 0.896 (95% CI, 0.797 to 0.957) for 'observation', respectively. CONCLUSIONS: The AI system achieved promising performance in the diagnosis of cataracts and corneal diseases.

来源:Br J Ophthalmol,查看原文。版权归原作者所有。

引用

  1. Ueno Y, Oda M, Yamaguchi T, et al. Deep learning model for extensive smartphone-based diagnosis and triage of cataracts and multiple corneal diseases. Br J Ophthalmol. 2024. PMID: 38242700. doi:10.1136/bjo-2023-324488

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。