Radiographer and PhD student
Sebastian Villesen Kristensen
Røntgen og Scanning, Kolding Sygehus
| Project management | ||
| Project status | Open | |
| Data collection dates | ||
| Start | 01.12.2024 | |
| End | 06.08.2025 | |
Optimisation of lower extremity angiography CT protocol using a noise reduction algorithm to evaluate PVAT in chronic and acute CLI patients
Short summary
The project sets out to investigate the effect of using different image reconstruction algorithms on PVAT Hounsfield units. With the introduction of AI-based algorithms to increase image quality the question of whether this affects Hounsfield units and therefore the diagnostic precision has to be raised. When filtered back projection was replaced by iterative reconstruction, changes had to be made, to prevent loss of image information. This is just as relevant in the case of AI-based software.
Rationale
Peripheral artery disease (PAD) is the third most common clinical manifestation of arteriosclerosis in patients, affecting about 236 million people worldwide1. PAD is highly related to age and the prevalence is expected to rise due to an ageing population and global increase in the prevalence of diabetes1,2. Mild symptoms of PAD can be fatigue, reduced mobility, pain and numbness when walking, severe symptoms are seen as critical limb ischemia (CLI), ulceration and gangrene ultimately resulting in amputation and death2,3. CLI patients have an expected one-year mortality rate of 15 to 40 %, while the 5-year mortality commonly exceeds 50 %, varying with more severe diseases 4. Post-amputation mortality rates increase, affecting the elderly patient to a higher extent 4. Atherosclerosis and its originating process "atherogenesis" has long been recognised as an inflammatory disease in the sub-intimal space, stemming from a response to injury5. Through multiple stages of chronic inflammation in the arterial wall, lesions will slowly advance and become stenotic5. However, pathophysiological studies have redirected the focus to adipose tissue surrounding the arterial wall. Perivascular adipose tissue (PVAT), has historically been thought to be a protective pad, with recent work showing, that the otherwise harmless tissue, possesses endocrine capabilities6. The proximity to the arterial wall has shown to be a factor in the formation of the atherosclerotic plaque, through the excretion of inflammatory mediators6-9. Atherosclerosis is initiated by stress-induced endothelial dysfunction and increased permeability of plasma constituents5,6. This is followed by smooth muscle cell migration and accumulation of oxidated low-density lipoprotein (LDL) in macrophage cells turning them into foam cells5,6. As foam cells and T-lymphocytes migrate into the plaque lumen, a fibrous cap forms creating a wall separating the lesion from the intimal lumen5,6. Lastly, the plaque may rupture in areas where the fibrous cap experiences continued flux of cell migration, leading to haemorrhaging from vasa vasorum or the arteria itself5. PVAT can assume both a pro- or anti-atherogenic effect with different adipokines having both protective and detrimental effects6,9. Since the recognition of PVAT's role in atherogenesis, a group at Oxford University has investigated the use of Computed Tomography (CT) for the detection of inflammation in coronary arteries10. This can be accomplished by a fixed x-ray tube and highly advanced detector rotation around the patient. The result of the data acquisition process is axial images showing the scanned anatomy on a standardized grey scale called Hounsfield Units (HU). It has been shown that HU values of PVAT can be used as a predictive measure for cardiac- and all-cause mortality, while also being able to quantify dynamic changes with increased arterial inflammation10,11. This can be achieved by monitoring changes in HU values with a less negative value constituting a change of increased aqueous phase in PVAT10,11. Currently, the main focus of PVAT is focused around cardiac CT, showing a predictive role in cardiac mortality8,10,12. However, the predictive role of PVAT in peripheral arteries, in the lower extremities (LE) is unknown. In present reports, vascular surgery in all Danish regions, amputation and preventive treatment of the lower extremity show, amputation being applied more often than preventive treatments13,14. This could be due to delayed patient admission resulting in late-stage progression in PAD. The hope is that vascular surgeons can use PVAT as a non-invasive predictor to follow the progression and intervene early on Study C - Optimisation of lower extremity angiography CT protocol using a noise reduction algorithm to evaluate PVAT in chronic and acute CLI patients. 1. A newly released noise reduction algorithm produces increased image quality compared to an iterative reconstruction algorithm. 2. Changes in algorithmic strength levels do not change the HU values of PVAT.
Description of the cohort
The projects cohort consist of patients suffering from chronic and acute CLI. They are included from department of Radiology at Lillebaelt Hospital Kolding. The categories of patients are either admitted as acute or outpatients.
Data and biological material
The main source of data in this particular project is Lower Extremity angiography CT scans.