Imaging Biomarkers – patient-friendly early warning systems
With imaging biomarkers, we use biological characteristics to monitor disease progression or support a diagnosis. The advantage is that they are usually non-invasive, do not require contrast media and can detect a disease even before it manifests.
Analysis methods supported by AI technologies
Today, mediri uses AI-based technology in various research projects on imaging biomarkers. This state-of-the-art technology helps to develop and refine analysis methods.
ASPIRE: Non-invasive Alzheimer’s early detection using ASL (Arterial Spin Labelling). Learn more
KI4MS: AI-assisted detection of smouldering lesions for follow-up of MS
BOSOMSHIELD: more precise breast cancer prognosis through comprehensive AI models. Learn more
Read more about the research projects in the slider below
Tracking Alzheimer’s non-invasively I – ASPIRE
Using arterial spin labelling (ASL), the blood flow in the brain is displayed non-invasively on MR images – without contrast media. As the head of a top-class European research consortium…
Non-invasively tracking Alzheimer’s II – ASPIRE
… we establish a standardised workflow that facilitates the evaluation and development of this first ASL-based Alzheimer’s imaging biomarker.
Image source: „ExploreASL: An image processing pipeline for multi-center ASL perfusion MRI studies” by Henk J.M.M. Mutsaerts et al. in NeuroImage (2020)
Multiple sclerosis in our sight
We develop and refine structural analysis methods to support the diagnosis and follow-up of multiple sclerosis.
Multiple Sclerosis in our sight – KI4MS
An important imaging biomarker for MS is smouldering lesions. Their AI-aided detection on MR images also makes disease activity visible that cannot be seen with the naked eye.
Image: KI4KMU – Example of an AI-supported VGM evaluation, which was developed as part of a project funded by the Baden-Württemberg Ministry of Economics together with the University Medical Center Mannheim and MedicalSyn GmbH in Stuttgart.
A comprehensive view of breast cancer – Bosomshield
The likelihood of breast cancer recurrence is still difficult to predict. Together with nine partners in the EU, we are the first to combine radiological, histo-pathological and clinical data of patients in one model to predict relapse of distant metastases.
« Biomarkers cover a broad spectrum: from body temperature and blood count to the expression of certain genes in cancer cells. MR-based biomarkers are particularly elegant because they are non-invasive, i.e., they do not require a biopsy or blood sample. »