Page 73 - ARNM-3-2

Advances in Radiotherapy
& Nuclear Medicine Modeling renal TAC in dynamic scintigraphy

(ii) 90 additional frames at one frame per minute, ensuring ensures consistency by using a rectangular
a complete observation of renal function over 30 min. selection over the kidneys. It is believed to
Step#2: The processing and conversion of signals: Since offer improved reproducibility by maintaining
the manually reconstructed TACs are based on gray-level standardized ROI dimensions across different
values, these images must undergo several transformation frames.
steps before they can be analyzed. The radioactive signals • Freehand method: Although not applied in
detected by the gamma camera are subjected to various this study, this alternative method allows for a
physical and digital processing operations aimed at more precise delineation of the kidneys’ actual
enhancing accuracy: shape. While offering greater flexibility, it has
(i) Gamma-ray detection: The radiation emitted by been suggested that it may introduce variability
the radiopharmaceutical is detected by the gamma depending on operator expertise.
camera. It has been noted that this detected signal (iv) Extraction of gray-level data: Following ROI selection,
consists of electrical pulses, which correspond to the gray-level values are extracted from the scintigraphic
intensity of gamma radiation. images. These values reflect the distribution of the
(ii) Amplification: To enhance sensitivity, the weak radioactive tracer within the kidneys over time.
gamma signal is amplified through a photomultiplier (v) Construction of TACs: The extracted gray-level data
tube. This component converts the low-energy gamma is used to manually reconstruct TACs, which describe
the uptake and clearance of the radiopharmaceutical
photons into stronger electrical signals.
(iii) Gain adjustment: The amplified signal is then over the 30-min imaging period. The manually
subjected to gain correction, a step which is intended generated TACs are then compared with the machine-
to optimize signal detection while reducing unwanted generated TACs, allowing an evaluation of their
accuracy in representing renal function.
variations. This adjustment ensures that the system
maintains an appropriate balance between sensitivity 2.3. Region of interest selection methods
and noise reduction.
(iv) Filtering: Various filtering techniques are applied to To study the impact of ROI selection methods in dynamic
refine the detected signal. Energy windowing and renal scintigraphy, two selection methods were compared:
scatter correction are commonly utilized to remove (i) Rectangular ROI selection: A fixed, standardized
background noise and irrelevant signals, thereby rectangular shape applied to selected data.
improving the reliability of the extracted data. (ii) Free-hand ROI selection: A manually drawn contour
(v) Analog-to-digital conversion: To allow for further following the anatomical shape of the kidneys.
digital analysis, the processed signals are converted The comparison was applied for six additional cases
into numerical pixel values. This conversion assigns (P1 – P6). The comparison of ROI selection methods was
intensity values to pixels in the generated images, performed to determine whether free-hand ROI selection
enabling subsequent image processing and data improves the accuracy of renal function assessment of the
extraction. considered cases in terms of kinetic parameters.
Step#3: The manual reconstruction of TACs: Once
the images are processed, manual TAC reconstruction is 2.4. TAC modeling using a one-compartment model
performed. The method follows a structured approach: and fitting function
(i) Extraction of scintigraphic images: The scintigraphic Most kinetic analyses in dynamic renal scintigraphy
images, stored in DICOM format, are retrieved from rely on mathematical models, typically compartmental
the gamma camera system for further analysis. models, to study the behavior of a tracer from injection
(ii) Image visualization and ROI selection: The micro- to clearance. These models incorporate an input function
DICOM software (v3.1.1) is employed to display that represents the amount of radioactivity injected into
and analyze the extracted images. The ROIs for both the patient. The proposed models take integro-differential
kidneys and background areas are then delineated, forms and are time-dependent. Such modeling can provide
ensuring that renal function is assessed in a controlled information and data on the dynamics of the tracer within
and reproducible manner. the patient and the target organ (the kidney) without the
(iii) Methods of ROI selection: Two main approaches for need for experiments or clinical trials. The estimation of
defining the ROIs: the kinetic parameters within the compartmental model
• Regular shape method (rectangular ROI): This is a non-linear regression that uses iterative algorithms,
approach, which was implemented in this study, based on the following assumptions: 13

Volume 3 Issue 2 (2025) 65 doi: 10.36922/ARNM025070008