Smart identification of psoriasis by images using convolutional neural networks: a case study in China

Abstract

Background Psoriasis is a chronic inflammatory skin disease, which holds a high incidence in China. However, professional dermatologists who can diagnose psoriasis early and correctly are insufficient in China, especially in the rural areas. A smart approach to identify psoriasis by pictures would be highly adaptable countrywide and could play a useful role in early diagnosis and regular treatment of psoriasis. Objectives Design and evaluation of a smart psoriasis identification system based on clinical images (without relying on a dermatoscope) that works effectively similar to a dermatologist. Methods A set of deep learning models using convolutional neural networks (CNNs) was explored and compared in the system for automatic identification of psoriasis. The work was carried out on a standardized dermatological dataset with 8021 clinical images of 9 common disorders including psoriasis along with full electronic medical records of patients built over the last 9 years in China. A two‐stage deep neural network was designed and developed to identify psoriasis. In the first stage, a multilabel classifier was trained to learn the visual patterns for each individual skin disease. In the second stage, the output of the first stage was utilized to distinguish psoriasis from other skin diseases. Results The area under the curve (AUC) of the two‐stage model reached 0.981 ± 0.015, which outperforms a single‐stage model. And, the classifier showed superior performance (missed diagnosis rate: 0.03, misdiagnosis rate: 0.04) than 25 Chinese dermatologists (missed diagnosis rate: 0.19, misdiagnosis rate: 0.10) in the diagnosis of psoriasis on 100 clinical images. Conclusions Using clinical images to identify psoriasis is feasible and effective based on CNNs, which also builds a solid technical base for smart care of skin diseases especially psoriasis using mobile/tablet applications for teledermatology in China.

A Deep Learning Based Framework for Diagnosing Multiple Skin Diseases in a Clinical Environment

“…In China, the board-certificated dermatologists are sparse, with the dermatologist to patient ratio as low as 1:60,000. Furthermore, the majority of well-trained dermatologists are practicing in large cities which worsens the reality that the demand for dermatological consultant is increasingly difficult to satisfy in remote and rural areas of China (23). Moreover, the capacity of giving correct diagnosis and management plans of Chinese dermatologists vary tremendously due to the imbalanced training and learning opportunities of FIGURE 6 | Confusion matrix of the classification result.…”

Machine Learning Applications in the Evaluation and Management of Psoriasis: A Systematic Review

“…We identified 8 articles that applied ML to identify an image of a psoriasis lesion as psoriasis and to differentiate it from other skin diseases. [21][22][23][24][25][26][27][28] Shrivastava et al have conducted a few studies to classify skin images from psoriasis patients as healthy versus diseased. After extracting feature information such as texture, color, and redness from images of psoriasis lesions, they used a support vector machine (SVM) model to classify 540 skin images from 30 psoriasis patients of Indian descent as healthy versus diseased, with a classification accuracy of approximately 99%.…”

“…[24][25][26] Other groups have focused on identifying psoriasis from images representing several common skin disorders, including diseases commonly mistaken for psoriasis like atopic dermatitis. [21][22][23]27,28 For example, Zhao et al used convolutional neural networks to classify 8021 images of 9 common disorders-lichen planus, lupus erythematosus, basal cell carcinoma, squamous cell carcinoma, atopic dermatitis, pemphigus, psoriasis, and seborrheic keratosis-from patients at a Chinese hospital as psoriasis versus non-psoriasis. 23 When tested on 100 new images, their algorithm showed superior performance to 25 Chinese dermatologists, with a misdiagnosis rate of 3% compared to 27% by dermatologists.…”

Künstliche Intelligenz zur Unterstützung der Telemedizin am Beispiel Afrikas

“…Künstliche Intelligenz (KI) hat sich zu einem relevanten Forschungsthema in der Medizin entwickelt und wird zunehmend in der Dermatologie angewandt. Die meisten KI-Anwendungen beschäftigten sich mit der Unterscheidung zwischen gutartigen und bösartigen Hautläsionen [8,9], andere mit entzündlichen Hautkrankheiten [10], der Allergologie [11] oder Dermatopathologie [12,13].…”

“…Künstliche Intelligenz (KI) hat sich zu einem relevanten Forschungsthema in der Medizin entwickelt und wird zunehmend in der Dermatologie angewandt. Die meisten KI-Anwendungen beschäftigten sich mit der Unterscheidung zwischen gutartigen und bösartigen Hautläsionen [ 8 , 9 ], andere mit entzündlichen Hautkrankheiten [ 10 ], der Allergologie [ 11 ] oder Dermatopathologie [ 12 , 13 ]. Obwohl die Zahl der Studien zunimmt, gibt es interessanterweise relativ wenige Arbeiten, bei denen Dermatologen an der Konzeption, Gestaltung und Interpretation der Studien maßgeblich beteiligt sind.…”