Neurovascular Imaging Research Group
Neurovascular Imaging Research Group
The research group of Prof. Dr. Andreas Harloff is part of the Department of Neurology and Neurophysiology of the University Hospital and the University of Freiburg. The primary objectives of our research are as follows:
1. Identification of novel biomarkers of plaque instability in carotid arteries and in the aorta. To this end, we utilize various high-resolution 3D MRI sequences to assess individual geometry and 3D blood flow. The identification of novel risk factors leading to progression, rupture and cerebral embolization of arteriosclerotic plaques is our major objective, for which additional blood biomarkers will also be used.
2. Therapy decision (surgery versus drug therapy) for carotid stenosis. The utilization of innovative MRI and blood biomarkers in conjunction with recent ipsilateral retinal or cerebral symptoms and stenosis degree facilitates the identification of individual risk of rupture in patients. This enables timely surgical intervention (e.g., in cases of plaque haemorrhage) or conservative treatment, as appropriate. The objective is to conduct a randomized-controlled study using such biomarkers.
3. Artificial intelligence for automated evaluation of 3D blood flow. The evaluation of blood flow is time-consuming and prone to errors if performed manually. We therefore evaluate 4D flow MRI data measured in vivo from two clinics (Chicago, Freiburg) in the carotid bifurcation and aorta. This allows us to swiftly evaluate previously unavailable parameters such as wall forces in the blood in large patient cohorts.
4. Clinical studies on acute therapy within the interdisciplinary neurovascular (INVAS) network. The network treats 9,000 acute stroke patients per year and performs thrombolysis in around 900 of these patients/year. Therefore, it is our intention to use this unique infrastructure to conduct prospective studies on new acute stroke therapies.
The following research projects are driven by the pursuit of these primary objectives:
- Three-dimensional analysis of hemodynamic wall parameters of the carotid bifurcation using computer-assisted 4D flow MRI (https://gepris.dfg.de/gepris/projekt/450764008).
The aim of this study is to examine 140 consecutive patients with 20-80% stenosis of the internal carotid artery (ICA) using innovative multimodal MRI and blood biomarkers. The additional analysis of 4D flow MRI data using computational fluid dynamics (CFD) will facilitate the investigation of regionally high-resolution wall forces and plaque pressure. Our objective is to identify parameters that are associated with rupture-prone plaques (AHA type VI lesions) or predict progression the stenosis in the 2-year follow-up.
We anticipate that the findings of this study will offer significant novel insights into the pathophysiology of ICA stenosis and the establishment of novel biomarkers that will enhance the risk assessment and individualized treatment of carotid stenosis.
2. Analysis of carotid artery bifurcation arteriosclerosis using 3D software, computational fluid dynamics and long-term follow-up gepris.dfg.de/gepris/projekt/298934133
The objective of this project is to analyze multimodal MRI data of patients with ≤50% ICA stenosis using software to determine plaque volume, AHA plaque classification and computational fluid dynamics (CFD) at 12-month follow-up. Furthermore, a long-term follow-up will be conducted after 7 years.
The overarching objective of this project is to investigate and characterize the biological effects of arteriosclerosis at the carotid bifurcation over an extended time period. Furthermore, the progression of arteriosclerosis will be quantitatively (plaque volume) and qualitatively (plaque composition) recorded.
3. Automated evaluation of 3D blood flow using AI methods
4D flow MRI allows in vivo measurement of individual blood flow in each vascular area of the body and the measurement of parameters that could not previously be measured, such as 3D wall forces or plaque pressure. The data is collected on various MRI devices, the analysis currently takes 30-60 minutes and is carried out using various evaluation methods. Consequently, the process is currently too time-consuming and not standardized enough for use in multicenter clinical studies. The aim of our study is to establish an automated analysis pipeline with extensive training data from manual segmentation of the aorta and carotid bifurcation at two locations (Chicago - Freiburg). This is based on artificial intelligence (AI) and neural networks and allows the data to be analyzed quickly and independently of the examiner. This methodology is a prerequisite for the use of 4D flow information in future multicenter studies.
4. Clinical stroke research in the interdisciplinary neurovascular network (INVAS)
With almost 10,000 stroke patients per year, the INVAS network offers excellent conditions for testing new diagnostic procedures, referral pathways or therapies in a homogeneous care infrastructure.
This network structure, located in southwestern Germany, is thus to be utilized for the initial evaluation of novel therapies, followed by their subsequent assessment through the standardized documentation of clinical parameters in subsequent follow-ups.