top of page

Artificial intelligence technology for remote assessment of eczema severity for diverse skin tones

Eczema

More Than Skin Disease

Eczema is an itchy skin condition that affects ~20% of UK schoolchildren. Constant itching and severe sleep loss may affect the quality of life for the whole family.

 

Timely and adequate treatment to get control of eczema and then keep control by preventing flares is critically important. Delayed medical consultations or undertreatment may result in worsening of symptoms.

00010-1463790609.png

"I need to go to clinics for my eczema severity to be assessed, as it is difficult to self-track eczema severity in a consistent and accurate way at home."

Eczema severity is a crucial piece of information for determining appropriate treatment. However, judging eczema severity based on digital images in teledermatology is challenging, particularly for darker skin.

Eczema severity assessment in darker skin can be difficult, even on face-to-face assessment, and requires specific training. Lack of experience and learning resources in assessing eczema severity in skin of colour leads to underestimation of eczema severity, resulting in a higher likelihood of under-treatment.

“We are experienced in assessing eczema severity but can be less experienced with assessing darker skin.”

EczemaNet

EczemaNet

A Breakthrough Solution

EczemaNet is an AI tool to robustly assess eczema severity objectively and remotely from digital images for children and young people of diverse skin types.

Pan et al 2020.png
Pan et al. 2020

Benefits

of EczemaNet for Healthcare Professionals & Patients

  • EczemaNet allows patients to assess their eczema severity at home and how their eczema severity varies over time in a simple and objective manner.

  • EczemaNet supports health care professionals with assessing eczema severity more quickly and accurately.

  • EczemaNet facilitates communication between patients and clinicians by providing a tool to demonstrate changes of the eczema condition over time.

  • EczemaNet provides a tool for remote assessment of clinical signs of eczema in clinical trials.

Our Research

There are currently two ongoing research projects for EczemaNet both funded by NIHR.
Find out more about our work to date

Leo Huang, Wai Hoh Tang, Rahman Attar, Claudia Gore, Hywel C. Williams, Adnan Custovic, Reiko J. Tanaka (2024)

"Remote assessment of eczema severity via AI-powered skin image analytics: A systematic review"

Artificial Intelligence in Medicine

​

R Attar, G Hurault, Z Wang, R Mokhtari, K Pan, B Olabi, E Earp, Lloyd Steele, HC Williams, RJ Tanaka (2023) 

"Reliable Detection of Eczema Areas for Fully Automated Assessment of Eczema Severity from Digital Camera Images"

JID Innovations

​

G Hurault, K Pan, R Mokhtari, B Olabi, E Earp, L Steele, HC Williams, RJ Tanaka (2022)

"Detecting Eczema Areas in Digital Images: An Impossible Task?"

JID Innovations

​

K Pan, G Hurault, K Arulkumaran, HC Williams, RJ Tanaka (2020)

"EczemaNet: Automating Detection and Severity Assessment of Atopic Dermatitis"

MLMI: International Workshop on Machine Learning in Medical Imaging

Research
About Us

About Us

We are committed to providing benefits for patients and clinicians and improving health outcomes in eczema.

NIHR logo 1.png

Meet the Team

Our team is ethnically diverse, gender-balanced and interdisciplinary.

Reiko Tanaka.jpg

Reiko Tanaka

Hywel Williams.jpg

Hywel Williams

Kim Thomas.jpg

Kim Thomas

James Moore Jr.jpg

James Moore Jr

Claudia Gore.jpg

Claudia Gore

Amanda Roberts.jpg

Amanda Roberts

Nadeem Qureshi.jpg

Nadeem Qureshi

Jena Strawford.jpg

Jena Strawford

Wai Hoh Tang.jpg

Wai Hoh Tang

Carron Layfield.jpg

Carron Layfield

Sharon Belmo.jpg

Sharon Belmo

Anne Roques.jpg

Anne Roques

Filip Paszkiewicz.jpg

Filip Paszkiewicz

Sabrina Kapur.jpg

Sabrina Kapur

template_edited.jpg

Teresa Tang

Our Partners

Nottingham 1.png
Reiko Tanaka.jpg

Reiko Tanaka

Contact
Contact Us

Get in Touch

EczemaNet team

Department of Bioengineering

Imperial College London

London SW7 2AZ

bottom of page