Jokerst Bioimaging Lab (UCSD) and PRL join forces to detect wounds

Symbolic picture for the article. The link opens the image in a large view.

BaCaTeC will fund a joint project between Jokerst Bioimaging Lab at UCSD and PRL with 9.000 EUR:

Ulcers damage skin and tissue in people and evolve to chronic wounds that are a major burden for the health-care system as well as the patients. More than 6.5M US citizens (DE: ~2M) are affected generating costs for the medical infrastructure of US$25B (DE: ~EUR4B) annually. In an earlier work, we showed that using a combination of photoacoustic and ultrasound imaging is capable of non-invasively capture chronic wounds. The detection itself has to be done by the naked eye, which is rather complex as the affected area could be widespread in the imaging is hybrid. The goal of our project is to jointly develop an artificial intelligence protoype capable of supporting the researcher/physician with detecting wounds or healthy segments within these data. In particular, we will use non-invasively acquired data of small animals and train a deep neural network to solve the automatic classification task.  To feed the network, expert annotations of high quality need to be made to guarantee high performance of the classifier and a valid evaluation. Neural networks have shown outstanding results in such image classification tasks in the medical domain and can receive FDA-approval. The results can serve as triaging method for wound assessment such that physicists get a list of acute cases to concentrate on as well as be used in home care environments. In a second step, the project shall be extended to human data as well as to staging of wounds.