Publication Details



Lower extremity peripheral arterial disease (PAD) is a cardiovascular disease affecting more than 230 million adults worldwide. The disease causes partial or complete narrowing of the arteries in the legs, reducing local blood flow. When combined with long-standing and poorly managed diabetes, PAD can lead to diabetic foot ulcers (DFU). If left untreated, these ulcers can lead to gangrenous tissue and eventually amputation. Enhancement of tissue perfusion around the ulcers by restoring the blood flow is vital to improve the healing of the ulcers. Until recently, intra-arterial digital subtraction angiography (IADSA) and selective arteriography served as the gold standard imaging techniques for examining PAD and DFU. However, due to their invasive nature, these techniques have been replaced by non-invasive modalities such as run-off computed tomography angiography (CTA) and magnetic resonance angiography (MRA). While CTA and MRA offer valuable morphological information on the vasculature, they do not provide adequate information on blood flow or tissue perfusion and do not allow for functional treatment monitoring. Recent developments in wide-beam CT scanners allow the acquisition of multiple three-dimensional CT volumes over the same vascular structure at a high temporal resolution. This technique also referred to as 4D-CT, allows for tracking contrast bolus through the arteries and soft tissue and could, as such, provide information on blood flow and tissue perfusion, which could potentially be applied in the evaluation of PAD and DFU. This thesis proposes a number of contributions to investigate the application of 4D-CTA for the evaluation of PAD and DFU and to facilitate the computation of clinically relevant parameters. In chapter 1, we provide a clinical background of PAD and DFU and the commonly applied examination modalities. In addition, we introduce the image processing techniques and hemodynamic background required to extract hemodynamic parameters. In chapter 2, we present a clinical feasibility study to quantify blood flow velocity in the below-the-knee (BTK) arteries in patients with suspected PAD using 4D-CT. To examine the accuracy and precision of blood flow velocity measurements using 4D-CT, we describe a technical feasibility study using phantom and repetitive measurements in chapter 3. The results from these chapters demonstrate that estimating the blood flow velocity in the BTK arteries is not only feasible but also precise using 4D-CT. In chapter 4, we propose an automated segmentation method to segment the BTK arteries. In addition, a novel spatio-temporal model is presented to fit the passing contrast bolus to extract multiple hemodynamic parameters. The reproducibility of this method is tested in terms of temporal resolution and spatial noise. In addition, the extracted hemodynamic parameters are correlated with the presence of PAD to examine the applicability of 4D-CT in its evaluation. The proposed method led to stable and physiologically plausible estimations of quantitative hemodynamic parameters, even in the case of stenotic arteries.* In chapter 5, we explored the applicability of the extracted hemodynamic parameters for a more detailed evaluation of PAD by correlating them with the presence and degree of the disease. Additionally, we developed a computer-aided diagnosis (CAD) tool to predict PAD on a patient level. This study demonstrates the potential clinical value of using 4D-CTA for PAD evaluation through the automated extraction of hemodynamic parameters. In chapter 6, we present the preliminary results of a case report investigating the applicability of 4D-CT for assessing tissue perfusion in DFU. We found that a combined CT angiography and perfusion protocol enables the assessment of three phases - arterial, perfusion, and venous - using minimal contrast (2 mL). In chapter 7, we apply the knowledge and processing techniques of the previous chapters to evaluate foot ulcers in patients with diabetes using 4D-CT. This exploratory study demonstrated the potential applicability of 4D-CT in DFU evaluation as both the proximal vasculature and the extent of the perfusion deficit in the microvasculature can be assessed. During this PhD research, we explored the feasibility of utilizing 4D-CTA for quantifying blood flow and tissue perfusion in evaluating PAD and DFU. Our research demonstrated that quantifying hemodynamic parameters using a balanced 4D-CTA protocol is feasible, allowing for the acquisition of dynamic 4D-CTA images with good spatial and temporal resolution while minimizing patient radiation exposure. An automated processing pipeline was developed to segment the BTK arteries and extract relevant modynamic parameters, yielding reproducible and robust measurements. The potential clinical value of these hemodynamic measurements has been examined and validated in clinical studies, demonstrating the feasibility of 4D-CTA in PAD and DFU evaluation. The contributions presented in this thesis serve as a foundation for further advancements in dynamic vascular and perfusion imaging and have the potential to enhance the diagnosis and management of PAD and DFU. When successfully integrated into the clinical workflow, 4D-CT could facilitate patient-specific diagnosis and treatment of PAD and DFU.