Introduction: Positron emission tomography (PET) is one of the most advanced methods in
modern imaging diagnostics, as it enables early detection of diseases by visualizing
metabolic and biochemical processes in the body. Compared to anatomical imaging
techniques such as CT and MRI, PET provides a functional insight into organ activity, which
is crucial for diagnostics in oncology, neurology, and cardiology. The accuracy and quality
of PET images depend on several parameters, among which the size of the image matrix is
a key factor. It affects spatial resolution, image noise, and the reliability of quantitative
measurements such as the standardized uptake value (SUV). Purpose: The purpose of this
thesis is to investigate, through a systematic literature review, the impact of image matrix
size on the quality and diagnostic value of PET images. The focus is on how matrix size
influences image reconstruction, spatial and contrast resolution, noise occurrence, and the
accuracy of standardized uptake value (SUV) measurements. Methods: In this thesis, we
used a descriptive method with a systematic review of scientific and professional literature.
We focused on studies conducted on phantoms or human subjects that examined the
influence of image matrix size in PET imaging. Literature was reviewed in both English and
Slovenian, primarily using databases such as Google Scholar, PubMed, ScienceDirect, and
official websites of professional institutions. Inclusion and exclusion criteria were applied,
with emphasis on thematic relevance, accessibility, and comprehensibility of the articles.
The literature review was conducted between December 2023 and May 2025. Results: A
systematic review of nine scientific studies examining the impact of image matrix size on
the quality and diagnostic utility of PET images. The analyses showed that increasing the
matrix size contributes to improved spatial resolution, higher SUV values, and enhanced
detectability of small lesions. However, this often comes at the cost of increased image noise
and greater variability in quantitative measurements, necessitating the use of appropriate
filtering and reconstruction methods such as PSF and BPL. Discussion and conclusion: The
analyzed studies demonstrated that larger matrices result in higher SUV values and better
visual detection but require careful interpretation due to the potential for false-positive
findings. Images with smaller voxels proved more reliable in quantifying small lesions while
maintaining diagnostic accuracy when properly filtered and processed. The matrix size
should always be tailored to the objective of the examination, the anatomical region, and the
capabilities of the system. A properly selected matrix, in combination with an appropriate
reconstruction method, enhances lesion detection, quantification, and overall quality of
patient care.
|
