Meng Yanda, Preston Frank George, Ferdousi Maryam, Azmi Shazli, Petropoulos Ioannis Nikolaos, Kaye Stephen, Malik Rayaz Ahmed, Alam Uazman, Zheng Yalin
Department of Eye and Vision Science, Institute of Life Course and Medical Sciences, University of Liverpool, Liverpool L7 8TX, UK.
Department of Cardiovascular & Metabolic Medicine, Institute of Life Course and Medical Sciences, University of Liverpool, Liverpool L7 8TX, UK.
J Clin Med. 2023 Feb 6;12(4):1284. doi: 10.3390/jcm12041284.
Diabetic peripheral neuropathy (DPN) is the leading cause of neuropathy worldwide resulting in excess morbidity and mortality. We aimed to develop an artificial intelligence deep learning algorithm to classify the presence or absence of peripheral neuropathy (PN) in participants with diabetes or pre-diabetes using corneal confocal microscopy (CCM) images of the sub-basal nerve plexus. A modified ResNet-50 model was trained to perform the binary classification of PN (PN+) versus no PN (PN-) based on the Toronto consensus criteria. A dataset of 279 participants (149 PN-, 130 PN+) was used to train ( = 200), validate ( = 18), and test ( = 61) the algorithm, utilizing one image per participant. The dataset consisted of participants with type 1 diabetes ( = 88), type 2 diabetes ( = 141), and pre-diabetes ( = 50). The algorithm was evaluated using diagnostic performance metrics and attribution-based methods (gradient-weighted class activation mapping (Grad-CAM) and Guided Grad-CAM). In detecting PN+, the AI-based DLA achieved a sensitivity of 0.91 (95%CI: 0.79-1.0), a specificity of 0.93 (95%CI: 0.83-1.0), and an area under the curve (AUC) of 0.95 (95%CI: 0.83-0.99). Our deep learning algorithm demonstrates excellent results for the diagnosis of PN using CCM. A large-scale prospective real-world study is required to validate its diagnostic efficacy prior to implementation in screening and diagnostic programmes.
糖尿病周围神经病变（DPN）是全球范围内神经病变的主要原因，会导致发病率和死亡率上升。我们旨在开发一种人工智能深度学习算法，使用角膜共聚焦显微镜（CCM）下基底神经丛图像，对糖尿病或糖尿病前期参与者的周围神经病变（PN）的有无进行分类。基于多伦多共识标准，训练了一个改进的ResNet-50模型，以对PN（PN+）与无PN（PN-）进行二元分类。使用279名参与者（149名PN-，130名PN+）的数据集训练（=200）、验证（=18）和测试（=61）该算法，每位参与者使用一张图像。该数据集包括1型糖尿病（=88）、2型糖尿病（=141）和糖尿病前期（=50）的参与者。使用诊断性能指标和基于归因的方法（梯度加权类激活映射（Grad-CAM）和引导式Grad-CAM）对该算法进行评估。在检测PN+时，基于人工智能的深度学习算法的灵敏度为0.91（95%CI：0.79-1.0），特异性为0.93（95%CI：0.83-1.0），曲线下面积（AUC）为0.95（95%CI：0.83-0.99）。我们的深度学习算法在使用CCM诊断PN方面显示出优异的结果。在将其应用于筛查和诊断项目之前，需要进行大规模前瞻性真实世界研究来验证其诊断效果。
