Deep Neural Network for Early Image Diagnosis of Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis
Deep Neural Network for Early Image Diagnosis of Stevens-Johnson Syndrome/Toxic Epidermal Necrolysis
复制标题
用于史蒂文斯-约翰逊综合征/中毒性表皮坏死松解症早期图像诊断的深度神经网络
DOI：
10.1016/j.jaip.2021.09.014
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Abe Riichiro
中科院分区：
文献类型：
--
作者：
Fujimoto Atsushi;Iwai Yuki;Ishikawa Takashi;Shinkuma Satoru;Shido Kosuke;Yamasaki Kenshi;Fujisawa Yasuhiro;Fujimoto Manabu;Muramatsu Shogo;Abe Riichiro
背景Stevens-Johnson综合征（SJS）/中毒性表皮坏死松解症（TEN）是一种危及生命的皮肤药物不良反应（cADR）。区分SJS/TEN从nonsevere cADR是困难的，特别是在疾病的早期阶段。ObjectiveTo克服这一局限性，我们开发了一个计算机辅助诊断系统，用于SJS/TEN的早期诊断，由深卷积神经网络（DCNN）提供动力。MethodsWe训练了DCNN使用的数据集的26,661个单独的病变图像从123例患者诊断SJS/TEN或nonsevere cADR。DCNN的分类准确率与10名委员会认证的皮肤科医生和24名实习皮肤科医生的分类准确率进行了比较。（95%置信区间[CI]，80.6-88.6），而委员会认证的皮肤科医生和实习皮肤科医生的敏感性为31.3%（95% CI，20.9-41.8;P< .0001）和27.8%（95% CI，22.6-32.5;P< .0001）。DCNN的阴性预测值为94.6%（95%CI，93.2-96.0），委员会认证的皮肤科医生为68.1%（95%CI，66.1-70.0;P< .0001），实习皮肤科医生为67.4%（95%CI，66.1-68.7;P< .0001）。接受者工作特征曲线下的DCNN为SJS/TEN诊断的面积为0.873，这是显着高于所有委员会认证的皮肤科医生和实习dermatologists.ConclusionsWe开发了一个DCNN分类SJS/TEN和nonsevere cADR的基础上，根据个人病变图像的红斑。DCNN在从皮肤图像分类SJS/TEN方面的表现明显优于皮肤科医生。
BackgroundStevens-Johnson syndrome (SJS)/toxic epidermal necrolysis (TEN) is a life-threatening cutaneous adverse drug reaction (cADR). Distinguishing SJS/TEN from nonsevere cADRs is difficult, especially in the early stages of the disease.ObjectiveTo overcome this limitation, we developed a computer-aided diagnosis system for the early diagnosis of SJS/TEN, powered by a deep convolutional neural network (DCNN).MethodsWe trained a DCNN using a dataset of 26,661 individual lesion images obtained from 123 patients with a diagnosis of SJS/TEN or nonsevere cADRs. The DCNN's accuracy of classification was compared with that of 10 board-certified dermatologists and 24 trainee dermatologists.ResultsThe DCNN achieved 84.6% sensitivity (95% confidence interval [CI], 80.6-88.6), whereas the sensitivities of the board-certified dermatologists and trainee dermatologists were 31.3 % (95% CI, 20.9-41.8;P< .0001) and 27.8% (95% CI, 22.6-32.5;P< .0001), respectively. The negative predictive value was 94.6% (95% CI, 93.2-96.0) for the DCNN, 68.1% (95% CI, 66.1-70.0;P< .0001) for the board-certified dermatologists, and 67.4% (95% CI, 66.1-68.7;P< .0001) for the trainee dermatologists. The area under the receiver operating characteristic curve of the DCNN for a SJS/TEN diagnosis was 0.873, which was significantly higher than that for all board-certified dermatologists and trainee dermatologists.ConclusionsWe developed a DCNN to classify SJS/TEN and nonsevere cADRs based on individual lesion images of erythema. The DCNN performed significantly better than did dermatologists in classifying SJS/TEN from skin images.