An interdisciplinary team from medicine and computer science has presented the app “Wunderkint” for the digital documentation and monitoring of chronic wounds. Those affected can use it to photograph the healing process themselves and transmit additional information such as pain intensity, itching or mood. The data is transmitted in encrypted form to treating specialists, who can track wound healing in real time via a dashboard and adjust the therapy if necessary. The aim is to improve care, increase the quality of life of patients and relieve the burden on the healthcare system.

The technical basis of the app is the AI system “WoundAmbit”, which was developed at the Chair of Software Engineering at the Julius-Maximilians-Universität Würzburg. The model uses methods of semantic segmentation to reliably recognize the wound area in photos and to determine the exact size and redness. For this purpose, reference cards with a colour scale and special markers are placed next to the wound to compensate for different recording conditions.

For the project, thousands of public wound images and a separate medical data set were evaluated. The results of the automatic analysis were compared with the assessments of experienced doctors to ensure the reliability of the technology. With this combination, it was possible to create a practical interface between modern AI and everyday care.

Vanessa Borst, PhD student at the Chair of Software Engineering, received the prize for the best student paper in the Applied Data Science Track at the European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD) 2025 in Porto for her work on the underlying paper. The project was convincing due to the successful bridging of highly developed semantic segmentation and its practical application in real wound care.

A feasibility study is currently underway, which shows initial positive results on the user-friendliness and acceptance of the app. In addition to telemedical care, appointments and video consultations can also be booked via the application. The developers see this as an important addition to classic wound care, which enables close observation without completely replacing personal contact with a doctor.

Original Paper:

[2506.06104] WoundAIssist: A Patient-Centered Mobile App for AI-Assisted Wound Care With Physicians in the Loop

Editor: X-Press Journalistenbüro GbR

Gender Notice. The personal designations used in this text always refer equally to female, male and diverse persons. Double/triple naming and gendered designations are used for better readability. ected.