Evolution of Computed Tomography

Abstract

This thesis investigates the clinical value of spectral CT for quantitative imaging and material-based diagnosis. The included studies address technical factors that influence quantitative measurements and evaluate the diagnostic performance of spectral CT reconstructions in different clinical applications. Part 1 – Measuring calcium Chapter 1.1 evaluates the influence of CT acquisition and reconstruction parameters on calcium quantification using conventional CT in peripheral arteries of the lower extremities. Both Agatston scoring and volumetric quantification were assessed. Reconstruction type and slice thickness significantly affected calcium measurements, whereas tube current and reader variability did not. These findings are clinically relevant because reconstruction-related variability may compromise longitudinal follow-up and comparisons between institutions, emphasizing the importance of standardized acquisition and reconstruction protocols. Chapter 1.2 investigates variability in calcium quantification across different CT scanners and scoring platforms using the same cadaver legs as in Chapter 1.1. Substantial variability was observed in Agatston scoring and volumetric measurements between conventional CT scanners, whereas differences between scoring platforms were limited. Scanner-related variability therefore remains an important source of inconsistency in conventional CT-based calcium quantification and may influence risk stratification or treatment decisions when patients are examined on different CT systems. Chapter 1.3 examines the added value of spectral CT material decomposition for measuring progression of brain calcifications in patients with Primary Familial Brain Calcification. Calcium mass measurements using material decomposition were accurate and appeared more sensitive for detecting progression than attenuation-based volumetric measurements and visual assessment. This increased sensitivity may improve monitoring of disease progression and evaluation of therapeutic effects. Part 2 – Measuring iodine Chapter 2.1 evaluates iodine quantification on portal venous dual-layer spectral CT for characterization of incidental indeterminate renal lesions. Using iodine concentration thresholds, non-enhancing lesions could be distinguished from enhancing lesions on portal venous phase CT alone. This approach may enable reliable identification of benign non-enhancing lesions and avoid additional imaging, thereby reducing patient anxiety, healthcare costs, and diagnostic burden. Chapter 2.2 assesses the diagnostic accuracy of dual-layer spectral CT for acute bowel ischemia in a large retrospective cohort. Adding spectral reconstructions did not improve sensitivity compared with conventional CT, although a small improvement in positive predictive value was observed with low monoenergetic reconstructions. Spectral CT may therefore reduce false-positive interpretations and unnecessary interventions, although further studies are needed to confirm this finding. Part 3 – Detection of bone lesions Chapter 3.1 presents a systematic review and meta-analysis comparing conventional CT, MRI, PET/CT, bone scintigraphy, and SPECT for detecting vertebral metastases. MRI showed the highest diagnostic accuracy on both patient and lesion level, whereas conventional CT had limited sensitivity. This highlights an important diagnostic gap because contrast-enhanced CT is routinely used for oncological staging and follow-up. Chapter 3.2 evaluates whether dual-layer spectral CT can improve detection of osteolytic vertebral metastases. Low monoenergetic reconstructions improved subjective lesion visibility, but diagnostic accuracy did not increase compared with conventional CT. This indicates that while spectral CT may improve lesion visibility during routine oncologic CT interpretation, it does not close the diagnostic gap identified in Chapter 3.1.