<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <title>OAR@UM Collection:</title>
  <link rel="alternate" href="https://www.um.edu.mt/library/oar/handle/123456789/150997" />
  <subtitle />
  <id>https://www.um.edu.mt/library/oar/handle/123456789/150997</id>
  <updated>2026-11-20T05:05:02Z</updated>
  <dc:date>2026-11-20T05:05:02Z</dc:date>
  <entry>
    <title>On the limitations of microwave reflectometry for microwave imaging in layered tissue phantoms</title>
    <link rel="alternate" href="https://www.um.edu.mt/library/oar/handle/123456789/151168" />
    <author>
      <name />
    </author>
    <id>https://www.um.edu.mt/library/oar/handle/123456789/151168</id>
    <updated>2026-11-18T14:12:56Z</updated>
    <published>2026-01-01T00:00:00Z</published>
    <summary type="text">Title: On the limitations of microwave reflectometry for microwave imaging in layered tissue phantoms
Abstract: Background: Microwave reflectometry is a non-ionising approach for detecting tissue abnormalities through dielectric-property contrast. However, its limitations in layered tissues, particularly the effects of skin thickness, tumour depth, frequency range, and tumour size on reflection-coefficient changes, remain insufficiently defined. Objectives: This study validated a numerical fat-phantom model against experimental measurements, investigated tumour depth and skin-thickness effects in layered tissue models, assessed whether extending the analysis to 50 GHz improved contrast resolution, and compared tumour-size-dependent responses under the most challenging simulated condition. Research Methodology: A fat-mimicking phantom was fabricated and measured using an open-ended coaxial probe and vector network analyser. The corresponding geometry was reproduced numerically and validated against the measured reflection response. The validated framework was then used to model layered skin–fat configurations containing spherical tumour inclusions. Detectability-related behaviour was assessed using the signed baseline-referenced difference ∆S11 = |S11,tumour|dB − |S11,baseline|dB, over 0.4–14 and 0.4–50 GHz, with |∆S11| used for magnitude-based interpretation. Results: The numerical model reproduced the main broadband trend of the measured fat-phantom response and was considered suitable for comparative analysis. Tumour-related |∆S11| responses generally decreased with increasing tumour depth and greater skin thickness in the lower-frequency range. Extending to 50 GHz increased observable contrast. In the most challenging condition, 8 mm and 9 mm tumours produced substantially larger responses than the 7 mm case and all smaller tumours. Conclusions and Recommendations: Microwave reflectometry in simplified layered tissue models is limited by superficial layer thickness, tumour depth, and weak reflected-signal perturbations. Higher-frequency analysis improved contrast resolution within the study assumptions, but detectability should be judged against setup-specific uncertainty. Future work should use layered physical phantoms, quantify full-band uncertainty, and investigate more realistic tissue models.
Description: B.Sc. (Hons)(Melit.)</summary>
    <dc:date>2026-01-01T00:00:00Z</dc:date>
  </entry>
  <entry>
    <title>The manufacturing and analysis of 3D-printed anthropomorphic organs for phantom studies</title>
    <link rel="alternate" href="https://www.um.edu.mt/library/oar/handle/123456789/151165" />
    <author>
      <name />
    </author>
    <id>https://www.um.edu.mt/library/oar/handle/123456789/151165</id>
    <updated>2026-11-18T14:04:40Z</updated>
    <published>2026-01-01T00:00:00Z</published>
    <summary type="text">Title: The manufacturing and analysis of 3D-printed anthropomorphic organs for phantom studies
Abstract: Background: The 3D printing process is delicate and complex, involving a variety of multidisciplinary techniques that require a steep learning curve to master. These techniques vary depending on the object being printed, and in medical physics research all prints must comply with specific criteria. Objectives: This dissertation aimed to compile a streamlined process of printing a selection of four diverse organ models, which were the brain, thyroid, liver and vertebrae. Applications for these models in medical physics research were also determined. It was hypothesised that smaller and more detailed organs would require special attention during segmentation and printing whereas more uniform organs would be easier to print. Research Methodology: The segmentation, refinement, slicing and printing processes were performed for all organ models, and the relevant numerical data and procedures were recorded using software packages such as 3D Slicer and PrusaSlicer. Quantitative data regarding print quality were recorded and used to amend the numerical parameters implemented in PrusaSlicer until a best fit was determined. Results: All models were designed and printed based on their use in medical physics research. Materials such as ABS, PETG and BVOH saw great success in creating the final version of the phantoms. Possible applications were provided for each phantom, such as in diagnostics, nuclear medicine, education and even in emerging areas such as clinical neurophysiology and surgery. Conclusions: 3D printing techniques are innovative technologies with a promising future in phantom manufacturing for medical physics research and emerging areas of healthcare. Despite the complexity of the process, 3D printing is a useful tool that can reduce financial burdens and contribute to personalised healthcare for treatments and procedures involving ionising and non-ionising radiation. Recommendations: Future work in this discipline could use different 3D printing techniques, such as SLA, further explore applications in radiation protection, and develop medical devices other than phantoms. Medical physicists should also understand the basics of 3D printing and device manufacturing to encourage wider use of locally produced custom medical devices at a lower cost.
Description: B.Sc. (Hons)(Melit.)</summary>
    <dc:date>2026-01-01T00:00:00Z</dc:date>
  </entry>
  <entry>
    <title>Radiation dose to fetus and breast of pregnant patients following CTPA and V/Q scans for suspected pulmonary embolism : a retrospective analysis in a national medical imaging department</title>
    <link rel="alternate" href="https://www.um.edu.mt/library/oar/handle/123456789/151162" />
    <author>
      <name />
    </author>
    <id>https://www.um.edu.mt/library/oar/handle/123456789/151162</id>
    <updated>2026-11-18T13:58:13Z</updated>
    <published>2026-01-01T00:00:00Z</published>
    <summary type="text">Title: Radiation dose to fetus and breast of pregnant patients following CTPA and V/Q scans for suspected pulmonary embolism : a retrospective analysis in a national medical imaging department
Abstract: Background: Pregnant patients investigated for suspected pulmonary embolism may undergo computed tomography pulmonary angiography or ventilation-perfusion imaging, both of which involve ionising radiation. Local fetal and maternal breast dose information was needed to support justified and optimised imaging decisions. Objectives: This study reviewed the literature on imaging for suspected pulmonary embolism in pregnancy, estimated and compared fetal doses from the two imaging approaches and estimated and compared maternal breast doses. Research Methodology: A quantitative retrospective comparative study was performed using anonymised examination data from pregnant patients investigated for suspected pulmonary embolism in a national Medical Imaging Department over ten years. Computed tomography pulmonary angiography doses were estimated using the National Cancer Institute computed tomography dosimetry software. Ventilation-perfusion doses were estimated using dose-coefficient calculations based on administered radiopharmaceutical activity and gestational stage. Dose estimates were compared descriptively by modality and gestation period, with exploratory statistical comparisons where sample size allowed. Results: The final dataset included 23 computed tomography pulmonary angiography cases and 7 planar perfusion imaging cases. Fetal absorbed dose estimates were low for both imaging approaches and all were below 1 mGy. Fetal dose estimates were generally lower for computed tomography pulmonary angiography than for perfusion or combined ventilation-perfusion imaging. Maternal breast absorbed dose was substantially higher for computed tomography pulmonary angiography, especially at 9 months’ gestation, where the mean dose was 5 ± 2 mGy compared with 0.25 ± 0.03 mGy for perfusion imaging and 0.52 ±0.03 mGy for combined ventilation-perfusion imaging. Conclusions and Recommendations: Modality selection should consider both fetal and maternal breast dose. Particular attention should be given to maternal breast dose, while computed tomography protocols and administered radiopharmaceutical activities should continue to be reviewed and optimised. Future work should use larger datasets, include diagnostic image quality and investigate scanner-specific dose optimisation.
Description: B.Sc. (Hons)(Melit.)</summary>
    <dc:date>2026-01-01T00:00:00Z</dc:date>
  </entry>
  <entry>
    <title>Optimising echo planar imaging pulse sequences for ultra-fast imaging on a 3 T MRI scanner</title>
    <link rel="alternate" href="https://www.um.edu.mt/library/oar/handle/123456789/151159" />
    <author>
      <name />
    </author>
    <id>https://www.um.edu.mt/library/oar/handle/123456789/151159</id>
    <updated>2026-11-18T13:49:26Z</updated>
    <published>2026-01-01T00:00:00Z</published>
    <summary type="text">Title: Optimising echo planar imaging pulse sequences for ultra-fast imaging on a 3 T MRI scanner
Abstract: Background: Echo Planar Imaging is a rapid magnetic resonance imaging technique used for functional magnetic resonance imaging, but at 3 Tesla it is vulnerable to geometric distortion, ghosting, signal loss and temporal instability. Objectives: This study aimed to determine how selected EPI parameter modifications affected phantom-based image quality, identify the main trade-offs and establish which fully evaluable modification showed the most favourable overall multi-metric response. Research Methodology: A quantitative experimental phantom-based design was used. A high-resolution T1-weighted reference image, a baseline Echo Planar Imaging acquisition and modified Echo Planar Imaging acquisitions with variations in acceleration, receiver bandwidth, echo time, field of view, flip angle, phase-encoding direction and distortion correction were obtained using one fluid-filled phantom. Image quality was assessed using geometric distortion, signal-to-noise ratio, ghosting ratio, signal uniformity and temporal stability. Protocol response was interpreted using a stated rule-based synthesis rather than a numerical composite score. Results: Echo time and flip angle produced the largest signal-related protocol effects. Exploratory testing indicated overall effects for geometric distortion, signal-to-noise ratio, ghosting ratio and signal uniformity. Short echo time gave the highest mean signal-to-noise ratio, increasing from 153 ± 8 at baseline to 180 ± 20, although this increase did not remain significant after Holm correction. Long echo time and low flip angle significantly reduced signal-to-noise ratio. Low flip angle improved signal uniformity to 73.7 ± 0.6% but reduced signal-to-noise ratio to 91 ± 3. Medium flip angle maintained signal-to-noise ratio close to baseline, improved signal uniformity to 71.6±0.9% and reduced ghosting to 0.23±0.07%, while geometric distortion remained below 1%. Ghosting and mean geometric distortion remained below 1% across all analysed protocols. The 200-volume time series showed limited temporal fluctuation. Conclusions and Recommendations: The tested EPI parameter modifications produced distinct image-quality trade-offs. Echo time and flip angle produced the largest signal-related effects. Under the stated rule-based criteria, medium flip angle showed the most favourable overall phantom-based response among the modifications. Further work should include re-&#xD;
peated scan sessions, additional scanners and volunteer or patient-based validation.
Description: B.Sc. (Hons)(Melit.)</summary>
    <dc:date>2026-01-01T00:00:00Z</dc:date>
  </entry>
</feed>

