Lu Jianyun, Tong Xiaoliang, Wu Hongping, Liu Yaoxinchuan, Ouyang Huidan, Zeng Qinghai
Department of Dermatology, Third Xiangya Hospital, Central South University, Changsha 410013, PR China.
Vocational Teachers College, Jiangxi Agricultural University, NanChang 330045, PR China.
Heliyon. 2023 Sep 16;9(9):e20186. doi: 10.1016/j.heliyon.2023.e20186. eCollection 2023 Sep.
Melasma (ML), naevus fusco-caeruleus zygomaticus (NZ), freckles (FC), cafe-au-lait spots (CS), nevus of ota (NO), and lentigo simplex (LS), are common skin diseases causing hyperpigmentation. Deep learning algorithms learn the inherent laws and representation levels of sample data and can analyze the internal details of the image and classify it objectively to be used for image diagnosis. However, deep learning algorithms that can assist clinicians in diagnosing skin hyperpigmentation conditions are lacking.
The optimal deep-learning image recognition algorithm was explored for the auxiliary diagnosis of hyperpigmented skin disease. Pretrained models, such as VGG-19, GoogLeNet, InceptionV3, ResNet50V2, ResNet101V2, ResNet152V2, InceptionResNetV2, DesseNet201, MobileNet, and NASNetMobile were used to classify images of six common hyperpigmented skin diseases. The best deep learning algorithm for developing an online clinical diagnosis system was selected by using accuracy and area under curve (AUC) as evaluation indicators.
In this research, the parameters of the above-mentioned ten deep learning algorithms were 18333510, 5979702, 21815078, 23577094, 42638854, 58343942, 54345958, 18333510, 3235014, and 4276058, respectively, and their training time was 380, 162, 199, 188, 315, 511, 471, 697, 101, and 144 min respectively. The respective accuracies of the training set were 85.94%, 99.72%, 99.61%, 99.52%, 99.52%, 98.84%, 99.61%, 99.13%, 99.52%, and 99.61%. The accuracy rates of the test set data were 73.28%, 57.40%, 70.04%, 71.48%, 68.23%, 71.11%, 71.84%, 73.28%, 70.39%, and 43.68%, respectively. Finally, the areas of AUC curves were 0.93, 0.86, 0.93, 0.91, 0.91, 0.92, 0.93, 0.92, 0.93, and 0.82, respectively.
The experimental parameters, training time, accuracy, and AUC of the above models suggest that MobileNet provides a good clinical application prospect in the auxiliary diagnosis of hyperpigmented skin.
黄褐斑（ML）、颧部褐青色痣（NZ）、雀斑（FC）、咖啡斑（CS）、太田痣（NO）和单纯性雀斑样痣（LS）是引起色素沉着的常见皮肤病。深度学习算法能够学习样本数据的内在规律和表示层次，可分析图像内部细节并进行客观分类，用于图像诊断。然而，目前缺乏能够辅助临床医生诊断皮肤色素沉着病症的深度学习算法。
探索用于色素沉着性皮肤病辅助诊断的最优深度学习图像识别算法。使用预训练模型，如VGG - 19、GoogLeNet、InceptionV3、ResNet50V2、ResNet101V2、ResNet152V2、InceptionResNetV2、DesseNet201、MobileNet和NASNetMobile对六种常见色素沉着性皮肤病的图像进行分类。以准确率和曲线下面积（AUC）作为评估指标，选择用于开发在线临床诊断系统的最佳深度学习算法。
本研究中，上述十种深度学习算法的参数分别为18333510、5979702、21815078、23577094、42638854、58343942、54345958、18333510、3235014和4276058，其训练时间分别为380、162、199、188、315、511、471、697、101和144分钟。训练集各自的准确率分别为85.94%、99.72%、99.61%、99.52%、99.52%、98.84%、99.61%、99.13%、99.52%和99.61%。测试集数据的准确率分别为73.28%、57.40%、70.04%、71.48%、68.23%、71.11%、71.84%、73.28%、70.39%和43.68%。最后，AUC曲线面积分别为0.93、0.86、0.93、0.91、0.91、0.92、0.93、0.92、0.93和0.82。
上述模型的实验参数、训练时间、准确率和AUC表明，MobileNet在色素沉着性皮肤的辅助诊断中具有良好的临床应用前景。
