{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T22:29:28Z","timestamp":1791325768622,"version":"4.3.1"},"reference-count":35,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2021,4,16]],"date-time":"2021-04-16T00:00:00Z","timestamp":1618531200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:p>\n                    <jats:bold>Background:<\/jats:bold>\n                    Numerous studies have attempted to apply artificial intelligence (AI) in the dermatological field, mainly on the classification and segmentation of various dermatoses. However, researches under real clinical settings are scarce.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Objectives:<\/jats:bold>\n                    This study was aimed to construct a novel framework based on deep learning trained by a dataset that represented the real clinical environment in a tertiary class hospital in China, for better adaptation of the AI application in clinical practice among Asian patients.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Methods:<\/jats:bold>\n                    Our dataset was composed of 13,603 dermatologist-labeled dermoscopic images, containing 14 categories of diseases, namely lichen planus (LP), rosacea (Rosa), viral warts (VW), acne vulgaris (AV), keloid and hypertrophic scar (KAHS), eczema and dermatitis (EAD), dermatofibroma (DF), seborrheic dermatitis (SD), seborrheic keratosis (SK), melanocytic nevus (MN), hemangioma (Hem), psoriasis (Pso), port wine stain (PWS), and basal cell carcinoma (BCC). In this study, we applied Google's EfficientNet-b4 with pre-trained weights on ImageNet as the backbone of our CNN architecture. The final fully-connected classification layer was replaced with 14 output neurons. We added seven auxiliary classifiers to each of the intermediate layer groups. The modified model was retrained with our dataset and implemented using Pytorch. We constructed saliency maps to visualize our network's attention area of input images for its prediction. To explore the visual characteristics of different clinical classes, we also examined the internal image features learned by the proposed framework using t-SNE (t-distributed Stochastic Neighbor Embedding).\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Results:<\/jats:bold>\n                    Test results showed that the proposed framework achieved a high level of classification performance with an overall accuracy of 0.948, a sensitivity of 0.934 and a specificity of 0.950. We also compared the performance of our algorithm with three most widely used CNN models which showed our model outperformed existing models with the highest area under curve (AUC) of 0.985. We further compared this model with 280 board-certificated dermatologists, and results showed a comparable performance level in an 8-class diagnostic task.\n                  <\/jats:p>\n                  <jats:p>\n                    <jats:bold>Conclusions:<\/jats:bold>\n                    The proposed framework retrained by the dataset that represented the real clinical environment in our department could accurately classify most common dermatoses that we encountered during outpatient practice including infectious and inflammatory dermatoses, benign and malignant cutaneous tumors.\n                  <\/jats:p>","DOI":"10.3389\/fmed.2021.626369","type":"journal-article","created":{"date-parts":[[2021,4,16]],"date-time":"2021-04-16T11:19:54Z","timestamp":1618571994000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":90,"title":["A Deep Learning Based Framework for Diagnosing Multiple Skin Diseases in a Clinical Environment"],"prefix":"10.3389","volume":"8","author":[{"given":"Chen-Yu","family":"Zhu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu-Kun","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hai-Peng","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kun-Lun","family":"Gao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chang","family":"Shu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun-Cheng","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Li-Feng","family":"Yan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi-Guang","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Feng-Ying","family":"Xie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jie","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2021,4,16]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"Litjens","year":"2017","journal-title":"Med Image Anal."},{"key":"B2","doi-asserted-by":"publisher","first-page":"1122","DOI":"10.1016\/j.cell.2018.02.010","article-title":"Identifying medical diagnoses and treatable diseases by image-based deep learning","volume":"172","author":"Kermany","year":"2018","journal-title":"Cell."},{"key":"B3","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1038\/nature21056","article-title":"Dermatologist-level classification of skin cancer with deep neural networks","volume":"542","author":"Esteva","year":"2017","journal-title":"Nature."},{"key":"B4","author":"Menegola","year":"2016","journal-title":"Towards automated melanoma screening: exploring transfer learning schemes. arXiv [Preprint]"},{"key":"B5","doi-asserted-by":"publisher","first-page":"795624","DOI":"10.1155\/2014\/795624","article-title":"SVM-RFE based feature selection and Taguchi parameters optimization for multiclass SVM classifier","volume":"2014","author":"Huang","year":"2014","journal-title":"Scient World J."},{"key":"B6","doi-asserted-by":"publisher","first-page":"1021","DOI":"10.1007\/s11548-017-1567-8","article-title":"Extraction of skin lesions from non-dermoscopic images for surgical excision of melanoma","volume":"12","author":"Jafari","year":"2017","journal-title":"Int J Comput Assist Radiol Surg."},{"key":"B7","doi-asserted-by":"publisher","first-page":"50","DOI":"10.1016\/j.compbiomed.2017.07.001","article-title":"Automatic discrimination of actinic keratoses from clinical photographs","volume":"88","author":"Spyridonos","year":"2017","journal-title":"Comput Biol Med."},{"key":"B8","doi-asserted-by":"publisher","first-page":"849","DOI":"10.1109\/TMI.2016.2633551","article-title":"Melanoma classification on dermoscopy images using a neural network ensemble model","volume":"36","author":"Xie","year":"2017","journal-title":"IEEE Trans Med Imaging."},{"key":"B9","doi-asserted-by":"publisher","first-page":"2245","DOI":"10.1007\/s11517-018-1835-y","article-title":"Classification of pressure ulcer tissues with 3D convolutional neural network","volume":"56","author":"Garcia-Zapirain","year":"2018","journal-title":"Med Biol Eng Comput."},{"key":"B10","doi-asserted-by":"publisher","first-page":"1529","DOI":"10.1016\/j.jid.2018.01.028","article-title":"Classification of the clinical images for benign and malignant cutaneous tumors using a deep learning algorithm","volume":"138","author":"Han","year":"2018","journal-title":"J Invest Dermatol."},{"key":"B11","doi-asserted-by":"publisher","first-page":"1305","DOI":"10.1111\/bjd.18741","article-title":"Predicting the long-term outcomes of biologics in patients with psoriasis using machine learning","volume":"182","author":"Emam","year":"2020","journal-title":"Br J Dermatol."},{"key":"B12","doi-asserted-by":"publisher","first-page":"373","DOI":"10.1111\/bjd.16924","article-title":"Deep-learning-based, computer-aided classifier developed with a small dataset of clinical images surpasses board-certified dermatologists in skin tumour diagnosis","volume":"180","author":"Fujisawa","year":"2019","journal-title":"Br J Dermatol."},{"key":"B13","doi-asserted-by":"publisher","first-page":"2027","DOI":"10.1097\/CM9.0000000000001023","article-title":"Deep learning-based, computer-aided classifier developed with dermoscopic images shows comparable performance to 164 dermatologists in cutaneous disease diagnosis in the Chinese population","volume":"133","author":"Wang","year":"2020","journal-title":"Chin Med J."},{"key":"B14","first-page":"191","article-title":"Applying a decision support system in clinical practice: results from melanoma diagnosis","volume":"2007","author":"Dreiseitl","year":"2007","journal-title":"AMIA Annu Symp Proc."},{"key":"B15","doi-asserted-by":"publisher","first-page":"180","DOI":"10.1097\/CMR.0b013e32832a1e41","article-title":"Computer versus human diagnosis of melanoma: evaluation of the feasibility of an automated diagnostic system in a prospective clinical trial","volume":"19","author":"Dreiseitl","year":"2009","journal-title":"Melanoma Res."},{"key":"B16","author":"Tan","year":"2019","journal-title":"EfficientNet: rethinking model scaling for convolutional neural networks. arXiv:1905.11946"},{"key":"B17","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1007\/s11263-015-0816-y","article-title":"ImageNet large scale visual recognition challenge","volume":"115","author":"Russakovsky","year":"2015","journal-title":"Int J Comp Vision."},{"key":"B18","doi-asserted-by":"publisher","first-page":"2011","DOI":"10.1109\/TPAMI.2019.2913372","article-title":"Squeeze-and-Excitation Networks","volume":"42","author":"Hu","year":"2020","journal-title":"IEEE Trans Pattern Anal Mach Intell."},{"key":"B19","first-page":"2818","volume-title":"Rethinking the inception architecture for computer vision","author":"Szegedy","year":"2015"},{"key":"B20","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1007\/s11263-019-01228-7","article-title":"Grad-CAM: visual explanations from deep networks via gradient-based localization","volume":"128","author":"Selvaraju","year":"2019","journal-title":"Int J Comp Vision."},{"key":"B21","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1109\/CVPR.2016.90","article-title":"Deep residual learning for image recognition","author":"He","year":"","journal-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"B22","first-page":"2579","article-title":"Visualizing high-dimensional data using t-SNE","volume":"9","author":"Hinton","year":"2008","journal-title":"J Mach Learn Res."},{"key":"B23","doi-asserted-by":"publisher","first-page":"518","DOI":"10.1111\/jdv.15965","article-title":"Smart identification of psoriasis by images using convolutional neural networks: a case study in China","volume":"34","author":"Zhao","year":"2020","journal-title":"J Eur Acad Dermatol Venereol."},{"key":"B24","doi-asserted-by":"publisher","first-page":"2119","DOI":"10.1097\/CM9.0000000000000389","article-title":"Assessment of imaging diagnosis ability of skin tumors in Chinese dermatologists","volume":"132","author":"Shen","year":"2019","journal-title":"Chin Med J (Engl)."},{"key":"B25","doi-asserted-by":"crossref","first-page":"246","DOI":"10.1023\/A:1018432632528","article-title":"Incidence of cutaneous melanoma among non-Hispanic whites, Hispanics, Asians, and blacks: an analysis of california cancer registry data, 1988-93","volume":"8","author":"Cress","year":"1997","journal-title":"Cancer Causes Control."},{"key":"B26","doi-asserted-by":"publisher","first-page":"vi1","DOI":"10.1093\/annonc\/mdp252","article-title":"Epidemiology of invasive cutaneous melanoma","volume":"20","author":"MacKie","year":"2009","journal-title":"Ann Oncol."},{"key":"B27","doi-asserted-by":"publisher","first-page":"232","DOI":"10.1111\/1346-8138.15640","article-title":"Dermoscopic diagnostic performance of Japanese dermatologists for skin tumors differs by patient origin: a deep learning convolutional neural network closes the gap","volume":"48","author":"Minagawa","year":"2020","journal-title":"J Dermatol"},{"key":"B28","doi-asserted-by":"publisher","first-page":"900","DOI":"10.1038\/s41591-020-0842-3","article-title":"A deep learning system for differential diagnosis of skin diseases","volume":"26","author":"Liu","year":"2020","journal-title":"Nat Med."},{"key":"B29","doi-asserted-by":"publisher","first-page":"169","DOI":"10.5826\/dpc.0903a01","article-title":"Dermoscopy of inflammatory dermatoses (Inflammoscopy): an up-to-date overview","volume":"9","author":"Errichetti","year":"2019","journal-title":"Dermatol Pract Concept."},{"key":"B30","doi-asserted-by":"publisher","first-page":"454","DOI":"10.1111\/bjd.18125","article-title":"Standardization of dermoscopic terminology and basic dermoscopic parameters to evaluate in general dermatology (non-neoplastic dermatoses): an expert consensus on behalf of the International Dermoscopy Society","volume":"182","author":"Errichetti","year":"2020","journal-title":"Br J Dermatol."},{"key":"B31","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1016\/j.ejca.2019.02.005","article-title":"A convolutional neural network trained with dermoscopic images performed on par with 145 dermatologists in a clinical melanoma image classification task","volume":"111","author":"Brinker","year":"2019","journal-title":"Eur J Cancer."},{"key":"B32","doi-asserted-by":"publisher","first-page":"1836","DOI":"10.1093\/annonc\/mdy520","article-title":"Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists","volume":"29","author":"Haenssle","year":"2018","journal-title":"Ann Oncol."},{"key":"B33","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1016\/j.ejca.2019.04.001","article-title":"Deep learning outperformed 136 of 157 dermatologists in a head-to-head dermoscopic melanoma image classification task","volume":"113","author":"Brinker","year":"2019","journal-title":"Eur J Cancer."},{"key":"B34","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1016\/j.ejca.2019.06.013","article-title":"Systematic outperformance of 112 dermatologists in multiclass skin cancer image classification by convolutional neural networks","volume":"119","author":"Maron","year":"2019","journal-title":"Eur J Cancer."},{"key":"B35","doi-asserted-by":"publisher","first-page":"1229","DOI":"10.1038\/s41591-020-0942-0","article-title":"Human-computer collaboration for skin cancer recognition","volume":"26","author":"Tschandl","year":"2020","journal-title":"Nat Med."}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2021.626369\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,4,16]],"date-time":"2021-04-16T11:20:04Z","timestamp":1618572004000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2021.626369\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,4,16]]},"references-count":35,"alternative-id":["10.3389\/fmed.2021.626369"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2021.626369","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,4,16]]},"article-number":"626369"}}