• Center on Health Equity and Access
  • Clinical
  • Health Care Cost
  • Health Care Delivery
  • Insurance
  • Policy
  • Technology
  • Value-Based Care

AI Smartphone App Could Improve Diagnosis of Psoriasis, Atopic Dermatitis, Eczema


Inflammatory skin diseases are often diagnosed by "first impression" and can be easily misdiagnosed. A new smartphone app using artificial intelligence (AI) may be able to assist with the diagnosis of psoriasis, atopic dermatitis, and eczema.

Inflammatory skin diseases affect more than 20% of the population worldwide and yet they are often diagnosed by “first impression,” and psoriasis, eczema, and atopic dermatitis are easily misdiagnosed.

Chinese researchers have developed an artificial intelligence dermatology diagnosis assistant (AIDDA) to diagnose psoriasis, eczema, and atopic dermatitis, and recognize healthy skin. According to the authors, convolutional neural networks (CNNs) have shown they can analyze clinical images and have been used to assist with the early diagnosis and detection of Alzheimer diseases and predict the risk of osteoarthritis from magnetic resonance imaging of knee cartilage, among other uses.

“There are now multiple examples of [artificial intelligence] tools facilitating cancer diagnosis based on data input from dermoscopes and from histological images of skin biopsy tissues,” the authors explained “However, to the best of our knowledge, we are unaware of any applications of AI tools to assist in diagnosing skin diseases other than cancers.”

The researchers collected skin images from people with healthy skin, psoriasis, eczema, and atopic dermatitis from the Second Xiangya Hospital. There were a total 4740 images, labeled by dermatologists in 3 categories: psoriasis, atopic dermatitis and eczema, or healthy.

The CNN model performed a diagnosis for each image among the 3 categories and it showed an accuracy of 95.80% [0.09%]. For psoriasis, the accuracy was 89.46% with a sensitivity of 91.40% and a specificity of 95.48%. The accuracy for atopic dermatitis and eczema was 92.57% with a sensitivity of 94.56% and specificity of 94.41%. The accuracy for healthy skin was even higher: 99.40% with sensitivity of 99.26% and specificity of 99.86%.

According to the authors, this study demonstrated that deep learning can effectively be applied in dermatology to differentiate amongst multiple diseases. The AIDDA smartphone app is publicly available to all doctors in China, and the authors noted that the app is already having an impact on the health care system there with more than 100,000 doctor-taken images being input already.

The authors pointed out that the tool will be of most help to inexperienced younger doctors and doctors in underdeveloped areas.

“Given that our system achieves a comparable or apparently superior performance to dermatologists for diagnosing inflammatory diseases, it seems obvious that smartphones enabled with deep learning network-developed apps will continue to benefit doctors in real-world clinical practice in dermatological and likely many other types of human disorders,” the authors concluded.


Wu H, Yin H, Chen H, et al. A deep learning, image based approach for automated diagnosis for inflammatory skin diseases. Ann Transl Med. 2020;8(9):581. doi:10.21037/atm.2020.04.39

Related Videos
Dr Michael Morse, Duke University
Video 10 - "Bronchiectasis Exacerbation Management"
video 10  - "Developing Practical Solutions to Improve Cardiovascular Care"
video 10  - "Developing Practical Solutions to Improve Cardiovascular Care"
Video 6 - "Optimal Approaches to Diagnostic Testing for Type 1 Diabetes Mellitus"
Video 13 - "Concluding Thoughts on Ph+ ALL Treatment"
Video 12 - "The Future of Ph+ ALL Treatments"
dr lawrence eichenfield
dr krystyn van vliet
Related Content
© 2024 MJH Life Sciences
All rights reserved.