An objective visualization and quantification of pain-generating processes in the periphery would alter pain diagnosis and represent an important paradigm shift in pain research. Positron emission tomography (PET) radioligand [11C]-D-deprenyl has shown an elevated uptake in painful inflammatory arthritis and whiplash-associated disorder. However, D-Deprenyl’s molecular binding target and uptake mechanism in inflammation and musculoskeletal injuries are still unknown. The present thesis aimed to gain insight into the mechanisms of D-deprenyl binding and uptake and to verify whether pain-associated sites and inflammation in acute musculoskeletal injury could be visualized, objectively quantified and followed over time with [11C]-D-deprenyl PET-computed tomography (PET/CT).

To identify the D-deprenyl binding target, a high-throughput analysis and competitive radioligand binding studies were performed. D-deprenyl inhibited monoamine oxidase A (MAO-A) activity by 55%, MAO-B activity by 99% and angiotensin-converting enzyme (ACE) by 70%, which identified these enzymes as higher-affinity targets. Furthermore, radioligand receptor binding assays pointed favorably towards the concept of MAO-B as the primary target. To investigate the biochemical characteristics of the binding site, we used radioligand binding assays to assess differences in the binding profile in inflamed human synovial membranes exhibiting varying levels of inflammation. D-deprenyl bound to a single, saturable population of membrane-bound protein in synovial membrane homogenates and the level of inflammation correlated with an increase in D-deprenyl binding affinity.

To verify whether D-deprenyl can visualize pain-generating processes, patients with musculoskeletal injuries were investigated and followed-up with [11C]-D-deprenyl PET/CT. In the study of eight patients with ankle sprain, the molecular aspects of inflammation and tissue injury could be visualized, objectively quantified and followed over time with [11C]-D-deprenyl PET/CT. The pain coexisted with increased [11C]-D-deprenyl uptake. In the study of 16 whiplash patients, an altered [11C]-D-deprenyl uptake in the cervical bone structures and facet joints was associated with subjective pain levels and self-rated disability.

To further evaluate D-Deprenyl’s usefulness as a marker of inflammation, three PET tracers were compared in an animal PET/CT study. Preliminary findings showed that [11C]-D-deprenyl had an almost identical uptake pattern when compared with [11C]-L-deprenyl. The two deprenyl enantiomers showed no signs of specific binding or trapping and therefore may not be useful to study further in models of inflammatory pain, surgical pain, or both.

This thesis demonstrates that D-deprenyl visualizes painful inflammation in musculoskeletal injuries and that the probable underlying mechanism of [11C]-D-deprenyl uptake is binding to MAO.

Purpose: The present study aims to analyze the accuracy of injections aimed to hit the proximal and depth part of the patellar tendon "target point" in patellar tendinopathy, comparing ultrasound-guided or non-ultrasound-guided (blind) injections.

Methods: A cadaver randomized study was carried out. Injections were performed under ultrasound control, as well as blinded. There were 26 knees from fresh cadavers and injections were placed by 26 practitioners with experience in the use of musculoskeletal ultrasound and injection treatment. Each participant performed 6 ultrasound-guided and 6 blind punctures in different cadaveric specimens. This provided 312 injections that were analyzed in 2 different anatomical cuts, thus providing a database of 624 measurements for statistical analysis.

Results: Statistically significant differences were observed (p < 0.0001) in the distance from the target point between the ultrasound-guided and the non-guided infiltrations. The "unguided" injections were considered to have been performed on average 10 mm away from the target point compared to the "ultrasound-guided" injections. The ultrasound-guided injections obtained an accuracy of 74.36% while the "non-ultrasound-guided" injections obtained an accuracy of 11.54% (p < 0.0001).

Conclusion: The use of ultrasound to guide the positioning of injections on the dorsal side of the proximal patellar tendon had a significantly higher accuracy compared to blind injections. The finding provides knowledge of importance for injection treatment.

Developing radiation delivery systems for optimisation of absorbed dose to the target without normal tissue toxicity requires advanced calculations for transport of radiation. In this thesis absorbed dose and fluence in different radiotherapy applications were calculated by using Monte Carlo (MC) simulations.

In paper I-III external neutron activation of gadolinium (Gd) for intravascular brachytherapy (GdNCB) and tumour therapy (GdNCT) was investigated. MC codes MCNP and GEANT4 were compared. MCNP was chosen for neutron capture reaction calculations. Gd neutron capture reaction includes both very short range (Auger electrons) and long range (IC electrons and gamma) products. In GdNCB the high-energetic gamma gives an almost flat absorbed dose delivery pattern, up to 4 mm around the stent. Dose distribution at the edges and inside the stent may prevent stent edge and in-stent restenosis. For GdNCT the absorbed dose from prompt gamma will dominate over the dose from IC and Auger electrons in an in vivo situation. The absorbed dose from IC electrons will enhance the total absorbed dose in the tumours and contribute to the cell killing.

In paper IV a model for calculation of inter-cluster cross-fire radiation dose from β-emitting radionuclides in a breast cancer model was developed. GEANT4 was used for obtaining absorbed dose. The dose internally in cells binding the isotope (self-dose) increased with decreasing β-energy except for the radionuclides with substantial amounts of conversion electrons and Auger electrons. An effective therapy approach may be a combination of radionuclides where the high self-dose from nuclides with low β-energy should be combined with the inter-cell cluster cross-fire dose from high energy β-particles.

In paper V MC simulations using correlated sampling together with importance sampling were used to calculate spectra perturbations in detector volumes caused by the detector silicon chip and its encapsulation. Penelope and EGSnrc were used and yielded similar results. The low energy part of the electron spectrum increased but to a less extent if the silicon detector was encapsulated in low z-materials.

BACKGROUND CONTEXT: The development of muscle fat infiltration (MFI) in the neck muscles is associated with poor functional recovery following whiplash injury. Custom software and time-consuming manual segmentation of magnetic resonance imaging (MRI) is required for quantitative analysis and presents as a barrier for clinical translation. PURPOSE: The purpose of this work was to establish a qualitative MRI measure for MFI and evaluate its ability to differentiate between individuals with severe whiplash-associated disorder (WAD), mild or moderate WAD, and healthy controls. STUDY DESIGN/SETTING: This is a cross-sectional study. PATIENT SAMPLE: Thirty-one subjects with WAD and 31 age-and sex-matched controls were recruited from an ongoing randomized controlled trial. OUTCOME MEASURES: The cervical multifidus was visually identified and segmented into eighths in the axial fat/water images (C4-C7). Muscle fat infiltration was assessed on a visual scale: 0 for no or marginal MFI, 1 for light MFI, and 2 for distinct MFI. The participants with WAD were divided in two groups: mild or moderate and severe based on Neck Disability Index % scores. METHODS: The mean regional MFI was compared between the healthy controls and each of the WAD groups using the Mann-Whitney U test. Receiver operator characteristic (ROC) analyses were carried out to evaluate the validity of the qualitative method. RESULTS: Twenty (65%) patients had mild or moderate disability and 11 (35%) were considered severe. Inter- and intra-rater reliability was excellent when grading was averaged by level or when frequency of grade II was considered. Statistically significant differences (pamp;lt;.05) in regional MFI were particularly notable between the severe WAD group and healthy controls. The ROC curve, based on detection of distinct MFI, showed an area-under-the curve of 0.768 (95% confidence interval 0.59-0.94) for discrimination of WAD participants. CONCLUSIONS: These preliminary results suggest a qualitative MRI measure for MFI is reliable and valid, and may prove useful toward the classification of WAD in radiology practice. (C) 2017 Elsevier Inc. All rights reserved.

The aims of this work were to image and analyze the needle behavior at automated core biopsy, to investigate the clinical utility of an alternative core biopsy technique using a semiautomated gun in breast and axillary lesions, and also to compare core biopsy with surgical specimens in malignant breast lesions regarding histologic features and hormone receptor expression.

In two experimental studies, using butter and silicon phantoms, respectively, the needle pass was imaged and its dynamic behavior studied. It was shown that the needle took a curved course in phantoms. It deviated to the same side as where the tip lay, and the degree of the curvature increased with increasing hardness of the phantoms. Our experimental methods can be applied for imaging of needle behavior and thereby improvement of needle configuration.

In two clinical studies, a semiautomated gun was used for large needle core biopsy of breast and axillary lesions in two series of 145 and 21 patients, respectively. The sensitivity of the method for diagnosis of malignancy was 87% (108/124), and in 37% (31/83) of cases the full length of the needle notch was filled with specimen. No injury to the neurovascular structures of the axillary area was observed. It was concluded that the semiautomated gun can be used as an alternative to the automated gun when the size and location of the lesion render use of the automatic device uncertain or dangerous, e.g., in small breast lesions or lesions located in the axilla.

In a series of 129 cases of breast cancer, comparison of core biopsy and surgical specimens showed that core biopsy provided enough information on the histologic type and grade of the lesions. Also, there was moderate to high concordance between the two methods for assessment of progesterone receptors and estrogen receptors (Spearman`s kappa 0.67 and 0.89, respectively).

In this document, technical details of the upgrade plan of the J-PARC neutrino beamline for the extension of the T2K experiment are described. T2K has proposed to accumulate data corresponding to 2×10²² protons-on-target in the next decade, aiming at an initial observation of CP violation with 3σ or higher significance in the case of maximal CP violation. Methods to increase the neutrino beam intensity, which are necessary to achieve the proposed data increase, are described.

Large-scale radiation emergency scenarios involving protracted low dose rate radiation exposure (e.g. a hidden radioactive source in a train) necessitate the development of high throughput methods for providing rapid individual dose estimates. During the RENEB (Running the European Network of Biodosimetry) 2019 exercise, four EDTA-blood samples were exposed to an Iridium-192 source (1.36 TBq, Tech-Ops 880 Sentinal) at varying distances and geometries. This resulted in protracted doses ranging between 0.2 and 2.4 Gy using dose rates of 1.5-40 mGy/min and exposure times of 1 or 2.5 h. Blood samples were exposed in thermo bottles that maintained temperatures between 39 and 27.7 degrees C. After exposure, EDTA-blood samples were transferred into PAXGene tubes to preserve RNA. RNA was isolated in one laboratory and aliquots of four blinded RNA were sent to another five teams for dose estimation based on gene expression changes. Using an X-ray machine, samples for two calibration curves (first: constant dose rate of 8.3 mGy/min and 0.5-8 h varying exposure times; second: varying dose rates of 0.5-8.3 mGy/min and 4 h exposure time) were generated for distribution. Assays were run in each laboratory according to locally established protocols using either a microarray platform (one team) or quantitative real-time PCR (qRT-PCR, five teams). The qRT-PCR measurements were highly reproducible with coefficient of variation below 15% in >= 75% of measurements resulting in reported dose estimates ranging between 0 and 0.5 Gy in all samples and in all laboratories. Up to twofold reductions in RNA copy numbers per degree Celsius relative to 37 degrees C were observed. However, when irradiating independent samples equivalent to the blinded samples but increasing the combined exposure and incubation time to 4 h at 37 degrees C, expected gene expression changes corresponding to the absorbed doses were observed. Clearly, time and an optimal temperature of 37 degrees C must be allowed for the biological response to manifest as gene expression changes prior to running the gene expression assay. In conclusion, dose reconstructions based on gene expression measurements are highly reproducible across different techniques, protocols and laboratories. Even a radiation dose of 0.25 Gy protracted over 4 h (1 mGy/min) can be identified. These results demonstrate the importance of the incubation conditions and time span between radiation exposure and measurements of gene expression changes when using this method in a field exercise or real emergency situation.

Early and high-throughput individual dose estimates are essential following large-scale radiation exposure events. In the context of the Running the European Network for Biodosimetry and Physical Dosimetry (RENEB) 2021 exercise, gene expression assays were conducted and their corresponding performance for dose-assessment is presented in this publication. Three blinded, coded whole blood samples from healthy donors were exposed to 0, 1.2 and 3.5 Gy X-ray doses (240 kVp, 1 Gy/min) using the X-ray source Yxlon. These exposures correspond to clinically relevant groups of unexposed, low dose (no severe acute health effects expected) and high dose exposed individuals (requiring early intensive medical health care). Samples were sent to eight teams for dose estimation and identification of clinically relevant groups. For quantitative reverse transcription polymerase chain reaction (qRT-PCR) and microarray analyses, samples were lysed, stored at 20°C and shipped on wet ice. RNA isolations and assays were run in each laboratory according to locally established protocols. The time-to-result for both rough early and more precise later reports has been documented where possible. Accuracy of dose estimates was calculated as the difference between estimated and reference doses for all doses (summed absolute difference, SAD) and by determining the number of correctly reported dose estimates that were defined as ±0.5 Gy for reference doses <2.5 Gy and ±1.0 Gy for reference doses >3 Gy, as recommended for triage dosimetry. We also examined the allocation of dose estimates to clinically/diagnostically relevant exposure groups. Altogether, 105 dose estimates were reported by the eight teams, and the earliest report times on dose categories and estimates were 5 h and 9 h, respectively. The coefficient of variation for 85% of all 436 qRT-PCR measurements did not exceed 10%. One team reported dose estimates that systematically deviated several-fold from reported dose estimates, and these outliers were excluded from further analysis. Teams employing a combination of several genes generated about two-times lower median SADs (0.8 Gy) compared to dose estimates based on single genes only (1.7 Gy). When considering the uncertainty intervals for triage dosimetry, dose estimates of all teams together were correctly reported in 100% of the 0 Gy, 50% of the 1.2 Gy and 50% of the 3.5 Gy exposed samples. The order of dose estimates (from lowest to highest) corresponding to three dose categories (unexposed, low dose and highest exposure) were correctly reported by all teams and all chosen genes or gene combinations. Furthermore, if teams reported no exposure or an exposure >3.5 Gy, it was always correctly allocated to the unexposed and the highly exposed group, while low exposed (1.2 Gy) samples sometimes could not be discriminated from highly (3.5 Gy) exposed samples. All teams used FDXR and 78.1% of correct dose estimates used FDXR as one of the predictors. Still, the accuracy of reported dose estimates based on FDXR differed considerably among teams with one team's SAD (0.5 Gy) being comparable to the dose accuracy employing a combination of genes. Using the workflow of this reference team, we performed additional experiments after the exercise on residual RNA and cDNA sent by six teams to the reference team. All samples were processed similarly with the intention to improve the accuracy of dose estimates when employing the same workflow. Re-evaluated dose estimates improved for half of the samples and worsened for the others. In conclusion, this inter-laboratory comparison exercise enabled (1) identification of technical problems and corrections in preparations for future events, (2) confirmed the early and high-throughput capabilities of gene expression, (3) emphasized different biodosimetry approaches using either only FDXR or a gene combination, (4) indicated some improvements in dose estimation with FDXR when employing a similar methodology, which requires further research for the final conclusion and (5) underlined the applicability of gene expression for identification of unexposed and highly exposed samples, supporting medical management in radiological or nuclear scenarios. 

The determination of the potential role and advantages of artificial intelligence-based models in the field of surgery remains uncertain. This research marks an initial stride towards creating a multimodal model, inspired by the Video-Audio-Text Transformer, that aims to reduce negative occurrences and enhance patient safety. The model employs text and image embedding state-of-the-art models (ViT and BERT) to assess their efficacy in extracting the hidden and distinct features from the surgery video frames. These features are then used as inputs for convolution-free Transformer architectures to extract comprehensive multidimensional representations. A joint space is then used to combine the text and image features extracted from both Transformer encoders. This joint space ensures that the relationships between the different modalities are preserved during the combination process. The entire model was trained and tested on laparoscopic cholecystectomy (LC) videos encompassing various levels of complexity. Experimentally, a mean accuracy of 91.0%, a precision of 81%, and a recall of 83% were reached by the model when tested on 30 videos out of 80 from the Cholec80 dataset.

Gastrin-releasing peptide receptors (GRPRs) are overexpressed in the majority of primary prostate tumors and in prostatic lymph node and bone metastases. Several GRPR antagonists were developed for SPECT and PET imaging of prostate cancer. We previously reported a preclinical evaluation of the GRPR antagonist [^99mTc]Tc-maSSS-PEG₂-RM26 (based on [D-Phe⁶, Sta¹³, Leu¹⁴-NH₂]BBN(6-14)) which bound to GRPR with high affinity and had a favorable biodistribution profile in tumor-bearing animal models. In this study, we aimed to prepare and test kits for prospective use in an early-phase clinical study. The kits were prepared to allow for a one-pot single-step radiolabeling with technetium-99m pertechnetate. The kit vials were tested for sterility and labeling efficacy. The radiolabeled by using the kit GRPR antagonist was evaluated in vitro for binding specificity to GRPR on PC-3 cells (GRPR-positive). In vivo, the toxicity of the kit constituents was evaluated in rats. The labeling efficacy of the kits stored at 4 °C was monitored for 18 months. The biological properties of [^99mTc]Tc-maSSS-PEG₂-RM26, which were obtained after this period, were examined both in vitro and in vivo. The one-pot (gluconic acid, ethylenediaminetetraacetic acid, stannous chloride, and maSSS-PEG₂-RM26) single-step radiolabeling with technetium-99m was successful with high radiochemical yields (>97%) and high molar activities (16–24 MBq/nmol). The radiolabeled peptide maintained its binding properties to GRPR. The kit constituents were sterile and non-toxic when tested in living subjects. In conclusion, the prepared kit is considered safe in animal models and can be further evaluated for use in clinics.

Background: Gastrin-releasing peptide receptor (GRPR) is an important target for imaging of prostate cancer. The wide availability of single-photon emission computed tomography/computed tomography (SPECT/CT) and the generator-produced ^99mTc can be utilized to facilitate the use of GRPR-targeting radiotracers for diagnostics of prostate cancers.

Methods: Synthetically produced mercaptoacetyl-Ser-Ser-Ser (maSSS)-PEG₂-RM26 and mercaptoacetyl-Ser-Glu-Ser (maSES)-PEG₂-RM26 (RM26 = d-Phe-Gln-Trp-Ala-Val-Gly-His-Sta-Leu-NH₂) were radiolabeled with ^99mTc and characterized in vitro using PC-3 cells and in vivo, using NMRI or PC-3 tumor bearing mice. SPECT/CT imaging and dosimetry calculations were performed for [^99mTc]Tc-maSSS-PEG₂-RM26.

Results: Peptides were radiolabeled with high yields (>98%), demonstrating GRPR specific binding and slow internalization in PC-3 cells. [^99mTc]Tc-maSSS-PEG₂-RM26 outperformed [^99mTc]Tc-maSES-PEG_2-RM26 in terms of GRPR affinity, with a lower dissociation constant (61 pM vs 849 pM) and demonstrating higher tumor uptake. [^99mTc]Tc-maSSS-PEG₂-RM26 had tumor-to-blood, tumor-to-muscle, and tumor-to-bone ratios of 97 ± 56, 188 ± 32, and 177 ± 79, respectively. SPECT/CT images of [^99mTc]Tc-maSSS-PEG₂-RM26 clearly visualized the GRPR-overexpressing tumors. The dosimetry estimated for [^99mTc]Tc-maSSS-PEG₂-RM26 showed the highest absorbed dose in the small intestine (1.65 × 10⁻³ mGy/MBq), and the effective dose is 3.49 × 10⁻³ mSv/MBq.

Conclusion: The GRPR antagonist maSSS-PEG₂-RM26 is a promising GRPR-targeting agent that can be radiolabeled through a single-step with the generator-produced ^99mTc and used for imaging of GRPR-expressing prostate cancer.

<bold>Introduction:</bold> Prostate specific membrane antigen (PSMA), highly expressed in metastatic castration-resistant prostate cancer (mCRPC), is an established therapeutic target. Theranostic PSMA-targeting agents are widely used in patient management and has shown improved outcomes for mCRPC patients. Earlier, we optimized a urea-based probe for radionuclide visualization of PSMA-expression in vivo using computer modeling. With the purpose to develop a targeting agent equally suitable for radionuclide imaging and therapy, the agent containing DOTA chelator was designed (BQ7876). The aim of the study was to test the hypothesis that Lu-177-labeled BQ7876 possesses target binding and biodistribution properties potentially enabling its use for radiotherapy.<bold>Methods:</bold> BQ7876 was synthesized and labeled with Lu-177. Specificity and affinity of [Lu-177]Lu-BQ7876 to PSMA-expressing PC3-pip cells was evaluated and its processing after binding to cells was studied. Animal studies in mice were performed to assess its biodistribution in vivo, target specificity and dosimetry. [Lu-177]Lu-PSMA-617 was simultaneously evaluated for comparison.<bold>Results:</bold> BQ7876 was labeled with Lu-177 with radiochemical yield >99%. Its binding to PSMA was specific in vitro and in vivo when tested in antigen saturation conditions as well as in PSMA-negative PC-3 tumors. The binding of [Lu-177]Lu-BQ7876 to living cells was characterized by rapid association, while the dissociation included a rapid and a slow phase with affinities K-D1 = 3.8 nM and K-D2 = 25 nM. The half-maximal inhibitory concentration for Lu-nat-BQ7876 was 59 nM that is equal to 61 nM for Lu-nat-PSMA-617. Cellular processing of [Lu-177]Lu-BQ7876 was accompanied by slow internalization. [Lu-177]Lu-BQ7876 was cleared from blood and normal tissues rapidly. Initial elevated uptake in kidneys decreased rapidly, and by 3 h post injection, the renal uptake (13 +/- 3%ID/g) did not differ significantly from tumor uptake (9 +/- 3%ID/g). Tumor uptake was stable between 1 and 3 h followed by a slow decline. The highest absorbed dose was in kidneys, followed by organs and tissues in abdomen.<bold>Discussion:</bold> Biodistribution studies in mice demonstrated that targeting properties of [Lu-177]Lu-BQ7876 are not inferior to properties of [Lu-177]Lu-PSMA-617, but do not offer any decisive advantages.

The targeting of gastrin-releasing peptide receptors (GRPR) was recently proposed for targeted therapy, e.g., radiotherapy. Multiple and frequent injections of peptide-based therapeutic agents would be required due to rapid blood clearance. By conjugation of the GRPR antagonist RM26 (D-Phe-Gln-Trp-Ala-Val-Gly-His-Sta-Leu-NH2) to an ABD (albumin-binding domain), we aimed to extend the blood circulation of peptides. The synthesized conjugate DOTA-ABD-RM26 was labelled with indium-111 and evaluated in vitro and in vivo. The labelled conjugate was stable in PBS and retained specificity and its antagonistic function against GRPR. The half-maximal inhibitory concentration (IC50) of In-nat-DOTA-ABD-RM26 in the presence of human serum albumin was 49 +/- 5 nM. [In-111]In-DOTA-ABD-RM26 had a significantly longer residence time in blood and in tumors (without a significant decrease of up to 144 h pi) than the parental RM26 peptide. We conclude that the ABD-RM26 conjugate can be used for GRPR-targeted therapy and delivery of cytotoxic drugs. However, the undesirable elevated activity uptake in kidneys abolishes its use for radionuclide therapy. This proof-of-principle study justified further optimization of the molecular design of the ABD-RM26 conjugate.

One novel option for treating metastatic castration resistant prostate cancer is radionuclide therapy targeting prostate-specific membrane antigen (PSMA), e.g. [Lu-177]Lu-PSMA-617. Overexpression of HER2 has been found in 80% of metastatic cases of prostate cancer. Previous research showed that HER2 is elevated post irradiation in PC-3 prostate cancer cells. Co-treating with anti-HER2 antibody Trastuzumab gave less proliferation of irradiated tumor cells in vitro, and when using radionuclide therapy, also in vivo. The aim of this study is to determine whether the same holds true in PSMA-expressing PC-3 PIP cells using [Lu-177]Lu-PSMA-617 radionuclide therapy. PC-3 PIP and 22Rv1 prostate cancer cells were tested in vitro, treated with 6 Gy of x-rays with or without Trastuzumab incubation. We measured uptake of HER2-targeting affibody [Ga-68]Ga-ABY-025 and cell survival, e.g. using the WST-1 assay. Three groups (n=10 each) of male nude Balb/c mice were inoculated with PC-3 PIP xenograft tumors and treated with just [Lu-177]Lu-PSMA-617 (20 MBq), [Lu-177]Lu-PSMA-617 (20 MBq) and Trastuzumab (4 x 5 mg/kg), or left untreated. Tumor sizes and animal survival was observed. In vitro, x-ray irradiation did reduce survival in 22Rv1 but not PC-3 PIP cells, and there was no significant effect of Trastuzumab treatment. Cells expressed HER2 but not significantly elevated post irradiation. In vivo, mice co-treated with Trastuzumab had significantly longer survival than untreated mice, but not than only [Lu-177]Lu-PSMA-617. Staining of tumor sections showed similar HER2 and PSMA expression across groups. In conclusion, these results give no support for any benefit from co-treatment with anti-HER2 antibody for PSMA-targeted radioligand therapy.

The bone-seeking radiopharmaceutical Xofigo (Radium-223 dichloride) has demonstrated both extended survival and palliative effects in treatment of bone metastases in patients with prostate cancer. The alpha-particle emitter Ra-223, administered as Ra-223 dichloride, targets regions undergoing active bone remodeling and strongly binds hydroxyapatite found in bone. However, the mechanisms mediating toxicity and properties of Ra-223 binding to hydroxyapatite are not fully understood. In the current study, we show that the alpha-particles originating from the Ra-223 decay chain produce a track-like distribution of the DNA damage response proteins 53BP1 and ɣH2AX and induce high amounts of clustered DNA double-strand breaks in prostate cancer cell nuclei. The Ra-223 treatment inhibited growth of prostate cancer cells, grown in 2D- and 3D- models in vitro, independent of prostate cancer cell type and androgen receptor variant 7 (ARv7) expression. The rapid binding with a high affinity of Ra-223 to bone structures was verified in an in silico assay (K_D= 19.2 ± 6.5 e-18) and almost no dissociation was detected within 24 hours. Importantly, there was no significant uptake of Ra-223 in cells. Further, we demonstrate the importance of the local dose-distribution of this treatment; there was more than 100-fold increase in cell killing when Ra-223 was attached to the bone-like hydroxyapatite structure, compared to when the radioactivity was distributed in the cell growth media. However, independent of the exposure condition, the high cell killing efficacy of the Ra-223 was attributed to the clustered DNA damaged sites induced by the released α-particles.

Background: Stereotactic radiosurgery (SRS) can be an effective primary or adjuvant treatment option for intracranial tumors. However, it carries risks of various radiation toxicities, which can lead to functional deficits for the patients. Current inverse planning algorithms for SRS provide an efficient way for sparing organs at risk (OARs) by setting maximum radiation dose constraints in the treatment planning process.Purpose: We propose using activation maps from functional MRI (fMRI) to map the eloquent regions of the brain and define functional OARs (fOARs) for Gamma Knife SRS treatment planning.Methods: We implemented a pipeline for analyzing patient fMRI data, generating fOARs from the resulting activation maps, and loading them onto the GammaPlan treatment planning software. We used the Lightning inverse planner to generate multiple treatment plans from open MRI data of five subjects, and evaluated the effects of incorporating the proposed fOARs.Results: The Lightning optimizer designs treatment plans with high conformity to the specified parameters. Setting maximum dose constraints on fOARs successfully limits the radiation dose incident on them, but can have a negative impact on treatment plan quality metrics. By masking out fOAR voxels surrounding the tumor target it is possible to achieve high quality treatment plans while controlling the radiation dose on fOARs.Conclusions: The proposed method can effectively reduce the radiation dose incident on the eloquent brain areas during Gamma Knife SRS of brain tumors.

Brain activation mapping using functional magnetic resonance imaging (fMRI) has been extensively studied in brain gray matter (GM), whereas in large disregarded for probing white matter (WM). This unbalanced treatment has been in part due to controversies in relation to the nature of the blood oxygenation level-dependent (BOLD) contrast in WM and its detachability. However, an accumulating body of studies has provided solid evidence of the functional significance of the BOLD signal in WM and has revealed that it exhibits anisotropic spatio-temporal correlations and structure-specific fluctuations concomitant with those of the cortical BOLD signal. In this work, we present an anisotropic spatial filtering scheme for smoothing fMRI data in WM that accounts for known spatial constraints on the BOLD signal in WM. In particular, the spatial correlation structure of the BOLD signal in WM is highly anisotropic and closely linked to local axonal structure in terms of shape and orientation, suggesting that isotropic Gaussian filters conventionally used for smoothing fMRI data are inadequate for denoising the BOLD signal in WM. The fundamental element in the proposed method is a graph-based description of WM that encodes the underlying anisotropy observed across WM, derived from diffusion-weighted MRI data. Based on this representation, and leveraging graph signal processing principles, we design subject-specific spatial filters that adapt to a subject’s unique WM structure at each position in the WM that they are applied at. We use the proposed filters to spatially smooth fMRI data in WM, as an alternative to the conventional practice of using isotropic Gaussian filters. We test the proposed filtering approach on two sets of simulated phantoms, showcasing its greater sensitivity and specificity for the detection of slender anisotropic activations, compared to that achieved with isotropic Gaussian filters. We also present WM activation mapping results on the Human Connectome Project’s 100-unrelated subject dataset, across seven functional tasks, showing that the proposed method enables the detection of streamline-like activations within axonal bundles.

Purpose: Clinical outcome after traumatic diffuse axonal injury (DAI) is difficult to predict. Three MRI techniques were compared in demonstrating acute brain lesions.  Relationship of the anatomical distribution of the lesions in combination with clinical prognostic factors to outcome after 6 months was evaluated.  

Methods and Materials: Thirty patients, aged 16-60 years (mean 31.2 years) with severe DAI (Glasgow Motor Score = GMS < 6) were examined with MRI at 1.5T within one week after the injury. A diffusion-weighted (DW) sequence (SE-EPI, b value 1000 s/mm²⁾, a T2*-weighted gradient echo (T2*GRE) sequence and a susceptibility-weighted (SWI) sequence were evaluated by two independent reviewers with short and long neuroradiological experiences. Clinical outcome was assessed with Extended Glasgow Outcome Score (GOSE) after ≥ 6 months.

Results: Interreviewer agreement for DAI classification was very good (ҡ 0.82 – 0.91) with all three sequences. SWI visualized more lesions than the T2*GRE or DW sequence.  In univariate analysis, number of DW lesions in the deep gray matter area including the internal capsules, number of SWI lesions in the mesencephalon, age, and GMS at admission and discharge correlated significantly with poor outcome.  Multivariate analysis only revealed an independent relation with poor outcome for age (p = 0.011) and lesions in the mesencephalic region including crura cerebri, substantia nigra and tegmentum on SWI (p = 0.032).

Conclusion: SWI is the most sensitive technique to visualize lesions in DAI. Age over 30 years and hemorrhagic mesencephalic lesions anterior to the tectum are indicators of poor long-term outcome in DAI.

Objective: Increased intracranial pressure (ICP) in patients with severe traumatic brain injury (TBI) with diffuse axonal injury (DAI) is not well defined. This study investigated the occurrence of increased ICP and whether clinical factors and lesion localization on MRI were associated with increased ICP in patients with DAI.

Methods: Fifty-two patients with severe TBI (median age 24 years, range 9–61 years), who had undergone ICP monitoring and had DAI on MRI, as determined using T2*-weighted gradient echo, susceptibility-weighted imaging, and diffusion-weighted imaging (DWI) sequences, were enrolled. The proportion of good monitoring time (GMT) with ICP > 20 mm Hg during the first 120 hours postinjury was calculated and associations with clinical and MRI-related factors were evaluated using linear regression.

Results: All patients had episodes of ICP > 20 mm Hg. The mean proportion of GMT with ICP > 20 mm Hg was 5%, and 27% of the patients (14/52) spent more than 5% of GMT with ICP > 20 mm Hg. The Glasgow Coma Scale motor score at admission (p = 0.04) and lesions on DWI sequences in the substantia nigra and mesencephalic tegmentum (SN-T, p = 0.001) were associated with the proportion of GMT with ICP > 20 mm Hg. In multivariable linear regression, lesions on DWI sequences in SN-T (8% of GMT with ICP > 20 mm Hg, 95% CI 3%–13%, p = 0.004) and young age (−0.2% of GMT with ICP > 20 mm Hg, 95% CI −0.07% to −0.3%, p = 0.002) were associated with increased ICP.

Conclusions: Increased ICP occurs in approximately one-third of patients with severe TBI who have DAI. Age and lesions on DWI sequences in the central mesencephalon (i.e., SN-T) are associated with elevated ICP. These findings suggest that MR lesion localization may aid prediction of increased ICP in patients with DAI.

Abbreviations: ADC = apparent diffusion coefficient; CPP = cerebral perfusion pressure; DAI = diffuse axonal injury; DWI = diffusion-weighted imaging; EVD = external ventricular drain; GCS = Glasgow Coma Scale; GMT = good monitoring time; GOSE = Glasgow Outcome Scale–Extended; ICC = intraclass correlation coefficient; ICP = intracranial pressure; MAP = mean arterial blood pressure; NICU = neurointensive care unit; SN-T = substantia nigra and mesencephalic tegmentum; SWI = susceptibility-weighted imaging; TBI = traumatic brain injury; T2*GRE = T2*-weighted gradient echo.

Objective: To evaluate whether artificial intelligence (AI) based automatic image analysis utilising convolutional neural networks (CNNs) can be used to evaluate computed tomography urography (CTU) for the presence of urinary bladder cancer (UBC) in patients with macroscopic hematuria. Methods: Our study included patients who had undergone evaluation for macroscopic hematuria. A CNN-based AI model was trained and validated on the CTUs included in the study on a dedicated research platform (Recomia.org). Sensitivity and specificity were calculated to assess the performance of the AI model. Cystoscopy findings were used as the reference method. Results: The training cohort comprised a total of 530 patients. Following the optimisation process, we developed the last version of our AI model. Subsequently, we utilised the model in the validation cohort which included an additional 400 patients (including 239 patients with UBC). The AI model had a sensitivity of 0.83 (95% confidence intervals [CI], 0.76-0.89), specificity of 0.76 (95% CI 0.67-0.84), and a negative predictive value (NPV) of 0.97 (95% CI 0.95-0.98). The majority of tumours in the false negative group (n = 24) were solitary (67%) and smaller than 1 cm (50%), with the majority of patients having cTaG1-2 (71%). Conclusions: We developed and tested an AI model for automatic image analysis of CTUs to detect UBC in patients with macroscopic hematuria. This model showed promising results with a high detection rate and excessive NPV. Further developments could lead to a decreased need for invasive investigations and prioritising patients with serious tumours.

The widespread use of medical imaging contributes to the increased detection of incidentally detected small renal tumors, a majority which are often indolent masses found in elderly patients with preexisting chronic kidney disease. In Sweden, partial nephrectomy with minimal invasive surgical approach is the current standard for removing these tumors, although another option is percutaneous image-guided tumor ablation that allows treatment of elderly patients with comorbidities for who surgery is a risk. Due to the lack of long-term follow-up studies and prospective randomized trials, ablation is still considered an alternative option to surgery in Sweden. The aim of this thesis was to evaluate treatment of T1 renal tumors with CT guided radiofrequency (RFA) and microwave ablation (MWA).

Factors affecting the efficacy rate of complete tumor ablation with RFA after a single session were evaluated (Paper I). Optimal electrode placement and a long tumor distance to the collecting system were associated with an increased primary efficacy. Renal tumor RFA was compared with laparoscopic partial nephrectomy (LPN: Papers II-III): both methods had comparable secondary efficacy rates, but RFA involved several treatment sessions. Total session times and hospitalization times were shorter and complications less frequent for RFA than for LPN (Paper II). After treatment, renal function impact was assessed by evaluation of both renal function quantity and quality through determination of the split renal function (SRF: Paper III). Standard renal function measurements were assessed and both RFA and LPN were nephron sparing when treating small renal tumors and did not affect creatinine or GFR. However, LPN involved greater SRF reduction in the affected kidney than RFA. Initial experience with microwave ablation was evaluated and this new ablation technique demonstrated high efficacy rates with fewer complications, and was comparable with the mid-term results of now established ablation techniques (Paper IV).

In conclusion, CT guided RFA and MWA are safe and effective treatments for the removal of T1 renal tumors. This thesis provides further insights into the field of thermal ablation of small renal masses, which can aid future treatment selection and patient management.

Background

Thermal ablation (TA) with radiofrequency (RFA) or cryoablation (CA) are established treatments for small renal masses (≤4 cm). Microwave ablation (MWA) has several potential benefits (decreased ablation time, less susceptibility to heat-sink, higher lesion temperatures than RFA) but is still considered experimental considering the available small-sample studies with short follow-up.

Purpose

To evaluate technique efficacy and complications of our initial experience of renal tumors treated using percutaneous MWA with a curative intent.

Material and Methods

A total of 105 renal tumors (in 93 patients) were treated between April 2014 and August 2017. MWA was performed percutaneously with computed tomography (CT) guidance under conscious sedation (n=82) or full anesthesia. Patients were followed with contrast-enhanced CT scans at six months and yearly thereafter for a minimum of five years. The mean follow-up time was 2.1 years. The percentage of tumors completely ablated in a single session (primary efficacy rate) and those successfully treated after repeat ablation (secondary efficacy rate) were recorded. Patient and tumor characteristics as well as complications were collected retrospectively.

Results

The median patient age was 70 years and median tumor size was 25 mm. Primary efficacy rate was 96.2% (101/105 tumors). After including two residual tumors for a second ablation session, secondary efficacy was 97.1% (102/105). Periprocedural complications were found in 5.2% (5/95) sessions: four Clavien-Dindo I and one Clavien-Dindo IIIa. One postprocedural Clavien-Dindo II complication was found.

Conclusion

MWA has high efficacy rates and few complications compared to other TA methods at a mean follow-up of two years.

Background: Comparable oncological outcomes have been seen after surgical nephrectomy and thermal ablation of renal tumors recently. However, periprocedural outcome needs to be assessed for aiding treatment decision.

Purpose: To compare efficacy rates and periprocedural outcome (technical success, session time, hospitalization time, and complications) after renal tumor treatment with laparoscopic partial nephrectomy (LPN) or radiofrequency ablation (RFA).

Material and Methods: The initial experience with 49 (treated with LPN) and 84 (treated with RFA) consecutive patients for a single renal tumor (diameter ≤ 5 cm, limited to the kidney) during 2007-2014 was evaluated. Patient and tumor characteristics, efficacy rates, and periprocedural outcome were collected retrospectively. The stratified Mantel Haenzel and Van Elteren tests, adjusted for tumor complexity (with the modified R.E.N.A.L nephrometry score [m-RNS]), were used to assess differences in treatment outcomes.

Results: Primary efficacy rate was 98% for LPN and 85.7% for RFA; secondary efficacy rate was 93.9% for LPN and 95.2% for RFA; and technical success rate was 87.8% for LPN and 100% for RFA. Median session (m-RNS adjusted P < 0.001; LPN 215 min, RFA 137 min) and median hospitalization time were longer after LPN (m-RNS adjusted P < 0.001; LPN 5 days, RFA 2 days). Side effects were uncommon (LPN 2%, RFA 4.8%). Complications were more frequent after LPN (m-RNS adjusted P < 0.001; LPN 42.9%, RFA 10.7%).

Conclusion: Both methods achieved equivalent secondary efficacy rates. RFA included several treatment sessions, but session and hospitalization times were shorter, and complications were less frequent than for LPN. The differences remained after adjustment for renal tumor complexity.

BACKGROUND: Radiofrequency ablation (RFA) can be used to treat renal masses in patients where surgery is preferably avoided. As tumor size and location can affect ablation results, procedural planning needs to identify these factors to limit treatment to a single session and increase ablation success.

PURPOSE: To identify factors that may affect the primary efficacy of complete renal tumor ablation with radiofrequency after a single session.

MATERIAL AND METHODS: Percutaneous RFA (using an impedance based system) was performed using computed tomography (CT) guidance. Fifty-two renal tumors (in 44 patients) were retrospectively studied (median follow-up, 7 months). Data collection included patient demographics, tumor data (modified Renal Nephrometry Score, histopathological diagnosis), RFA treatment data (electrode placement), and follow-up results (tumor relapse). Data were analyzed through generalized estimating equations.

RESULTS: Primary efficacy rate was 83%. Predictors for complete ablation were optimal electrode placement (P = 0.002, OR = 16.67) and increasing distance to the collecting system (P = 0.02, OR = 1.18). Tumor size was not a predictor for complete ablation (median size, 24 mm; P = 0.069, OR = 0.47), but all tumors ≤2 cm were completely ablated. All papillary tumors and oncocytomas were completely ablated in a single session; the most common incompletely ablated tumor type was clear cell carcinoma (6 of 9).

CONCLUSION: Optimal electrode placement and a long distance from the collecting system are associated with an increased primary efficacy of renal tumor RFA. These variables need to be considered to increase primary ablation success. Further studies are needed to evaluate the effect of RFA on histopathologically different renal tumors.

Purpose Preclinical, or asymptomatic, Alzheimer's disease (AD) refers to the presence of positive AD biomarkers in the absence of cognitive deficits. This research concept is being applied to define target populations for clinical drug development. In a prospective community-recruited cohort of cognitively intact older adults, we compared two amyloid imaging markers within subjects: F-18-flutemetamol and C-11-Pittsburgh compound B (PIB). Methods In 32 community-recruited cognitively intact older adults aged between 65 and 80 years, we determined the concordance between binary classification based on F-18-flutemetamol versus C-11-PIB according to semiquantitative assessment (standardized uptake value ratio in composite cortical volume, SUVRcomp) and, alternatively, according to visual reads. We also determined the correlation between F-18-flutemetamol and C-11-PIB SUVR and evaluated how this was affected by the reference region chosen (cerebellar grey matter versus pons) and the use of partial volume correction (PVC) in this population. Results Binary classification based on semiquantitative assessment was concordant between F-18-flutemetamol and C-11-PIB in 94 % of cases. Concordance of blinded binary visual reads between tracers was 84 %. The Spearman correlation between F-18-flutemetamol and C-11-PIB SUVRcomp with cerebellar grey matter as reference region was 0.84, with a slope of 0.98. Correlations in neocortical regions were significantly lower with the pons as reference region. PVC improved the correlation in striatum and medial temporal cortex. Conclusion For the definition of preclinical AD based on F-18-flutemetamol, concordance with C-11-PIB was highest using semiquantitative assessment with cerebellar grey matter as reference region.

The relevance of gas-phase protein structure to its solution structure is of the utmost importance in studying biomolecules by mass spectrometry. D-Amino acid substitutions within a minimal protein. Trp-cage. were used to correlate solution-phase properties as measured by circular dichroism with solution/gas-phase conformational features of protein cations probed via charge state distribution (CSD) in electrospray ionization. and gas-phase features revealed by tandem mass spectrometry (MS/MS). The gas-phase features were additionally supported by force-field molecular dynamics simulations. CD data showed that almost any single-residue D-substitution destroys the most prominent CD feature of the "native" all-L isomer, alpha-helicity. CSD was able to qualitatively assess the degree of compactness of solution-phase molecular structures. CSD results were consistent with the all-L form being the most compact in solution among all studied stereoisomers except for the D-Asn(1) isomer. D-substitutions of the aromatic Y(3), W(6) and Q(5) residues generated the largest deviations in CSD data among single amino acid substitutions. consistent with the critical role of these residues in Trp-cage stability. Electron capture dissociation of the stereoisomer dications gave an indication that some gas-phase structural features of Trp-cage are similar to those in solution. This result is supported by MDS data oil five of the studied stereoisomer dications in the gas-phase. The MDS-derived minimum-energy structures possessed more extensive hydrogen bonding than the solution-phase structure of the native form, deviating from the latter by 3-4 angstrom and were not 'inside-out' compared to native structures. MDS data could be correlated with CD data and even with ECD results. which aided in providing a long-range structural constraint for MDS. The overall conclusion is the general resemblance, despite the difference on the detailed level, of the preferred structures in both phases for the mini protein Trp-cage.

Background: Radiobiological models for normal tissue complication probability (NTCP) are more and more commonly used in order to estimate the clinical outcome of radiation therapy. A normal tissue complication probability model to be considered a good and reliable one should fulfill the following two requirements: (a) it should predict the sigmoid shape of the dose-response curve as well as possible and (b) it should duly handle the volume effect. In the work from 2005 (IJROBP 61(3):892-900, 2005) P. van Luijk et al. suggest that none of the existing NTCP models is able to describe the volume effects present in the rat spinal cord during irradiation with small proton beams and they indicate the need for developing such new models.

Methods: We have used the experimental data from H. Bijl et al. (IJROBP 52(1):205-211, 2002) to try explaining the change in the fifty percent effective dose (ED50) for different field sizes. We initiated this study to evaluate whether the induction of white matter necrosis in rat spinal cord after irradiation with small proton beams could be explained independent of used NTCP model. We therefore introduced a new concept of effective FSU dose, where a convolution of the original dose distribution with a function describing the effective size of a single FSU results in the average doses in a functional subunit. Such procedure allows determining the ED50 in an FSU of a certain size, within the irradiation field. We have also looked at non uniform dose distributions to see whether using a similar method we can explain the so called “bath and shower experiments” (IJROBP 57(1): 274-281, 2003).

Results: Using the least square method to compare the effective doses for different sizes of functional subunits with the experimental data we observe the best fit for about 8 mm length. It seems that this length could be understood as an effective size of functional subunits in rat spinal cord, explaining what is otherwise interpreted as a volume effect. For the non uniform dose distributions an effective FSU length of 5 mm gives the optimal fit with the Probit dose-response model.

Conclusions: The concept of an effective FSU length seems to explain at least part of the effects seen when small portions of the rat spinal cord are irradiated. The most likely FSU length for the shower and bath experiments is 5 mm according to these calculations.

Bakgrund: Magnetresonanstomografi, MR, är en bilddiagnostisk undersökningsmetod som har ökat markant de senaste åren. Metoden kräver att patienten ligger stilla i ett trångt utrymme under längre tid, något som många kan uppleva oro och ångest inför. Detta leder till avbrutna och fördröjda undersökningar vilket kan resultera i mer stress för röntgensjuksköterskor samt ökad kostnad för samhället. 

Syfte: Syftet med studien är att undersöka hur oro och ångest kopplad till magnetkameraundersökningar skiljer sig emellan två sjukhus, samt vad de bakomliggande orsakerna är. Därtill ska patientupplevelsen av röntgensjuksköterskornas bemötande undersökas.

Metod: Studien var av empirisk kvantitativ design. Totalt 100 egenkonstruerade enkäter fördelades mellan Akademiska sjukhuset i Uppsala och Lasarettet i Enköping. Patienter som uppfyllde inklusionskriterierna tillfrågades om att anonymt deltaga i studien efter genomförd MR-undersökning.

Resultat: Den största faktorn till oro visade sig vara svaret på undersökningen. Andra vanliga faktorer till oro var att behöva ligga stilla samt det trånga utrymmet i MR-kameran. Patienternas upplevelse av information och bemötande från röntgensjuksköterskan visade att den absoluta majoriteten blev lugnare eller upplevde ingen skillnad i oro innan samt under undersökningen. Generellt var patienterna mycket nöjda med bemötandet vilket framgick i den kvalitativa analysen. Ingen signifikant skillnad upptäcktes mellan de båda sjukhusen i den jämförande analysen (p>0,05).

Slutsats: Svarsresultaten vid respektive sjukhus var väldigt liknande och ingen signifikant skillnad uppmättes i den jämförande analysen. Genomgående var att deltagarna blev lugnare av informationen och bemötandet som gavs. Studiens resultat kan bidra till ökad förståelse för patienters eventuella oro vid MR-undersökningar. 

The use of Magnetic Resonance Imaging (MRI) in the radiotherapy (RT) treatment planning workflow is increasing. MRI offers superior soft-tissue contrast compared to Computed Tomography (CT) and therefore improves the accuracy in target volume definitions. There are, however concerns with inherent geometric distortions from system- (gradient nonlinearities and main magnetic field inhomogeneities) and patient-related sources (magnetic susceptibility effect and chemical shift). The lack of clearly defined quality assurance (QA) procedures has also raised questions on the ability of current QA protocols to detect common image quality degradations under radiotherapy settings. To fully implement and take advantage of the benefits of MRI in radiotherapy, these concerns need to be addressed.

In Papers I and II, the dosimetric impact of MR distortions was investigated. Patient CTs (CT) were deformed with MR distortion vector fields (from the residual system distortions after correcting for gradient nonlinearities and patient-induced susceptibility distortions) to create distorted CT (dCT) images. Field parameters from volumetric modulated arc therapy (VMAT) treatment plans initially optimized on dCT data sets were transferred to CT data to compute new treatment plans. Data from 19 prostate and 21 head and neck patients were used for the treatment planning. The dCT and CT treatment plans were compared to determine the impact of distortions on dose distributions. No clinically relevant dose differences between distorted CT and original CT treatment plans were found. Mean dose differences were < 1.0% and < 0.5% at the planning target volume (PTV) for the head and neck, and prostate treatment plans, respectively. 

Strategies to reduce geometric distortions were also evaluated in Papers I and II. Using the vendor-supplied gradient non-linearity correction algorithm reduced overall distortions to less than half of the original value. A high acquisition bandwidth of 488 Hz/pixel (Paper I) and 488 Hz/mm (Paper II) kept the mean geometric distortions at the delineated structures below 1 mm. Furthermore, a patient-specific active shimming method implemented in Paper II significantly reduced the number of voxels with distortion shifts > 2 mm from 15.4% to 2.0%.

B₀ maps from patient-induced magnetic field inhomogeneities obtained through direct measurements and by simulations that used MR-generated synthetic CT (sCT) data were compared in Paper III. The validation showed excellent agreement between the simulated and measured B₀ maps.

In Paper IV, the ability of current QA methods to detect common MR image quality degradations under radiotherapy settings were investigated. By evaluating key image quality parameters, the QA protocols were found to be sensitive to some of the introduced degradations. However, image quality issues such as those caused by RF coil failures could not be adequately detected.

In conclusion, this work has shown the feasibility of using MRI data for radiotherapy treatment planning as distortions resulted in a dose difference of less than 1% between distorted and undistorted images. The simulation software can be used to produce accurate B₀ maps, which could then be used as the basis for the effective correction of patient-induced field inhomogeneity distortions and for the QA verification of sCT data. Furthermore, the analysis of the strengths and weaknesses in current QA tools for MRI in RT contribute to finding better methods to efficiently identify image quality errors.

Magnetic resonance imaging (MRI) utilizes the magnetic properties of tissues to generate image-forming signals. MRI has exquisite soft-tissue contrast and since tumors are mainly soft-tissues, it offers improved delineation of the target volume and nearby organs at risk. The proposed Magnetic Resonance-only Radiotherapy (MR-only RT) work flow allows for the use of MRI as the sole imaging modality in the radiotherapy (RT) treatment planning of cancer. There are, however, issues with geometric distortions inherent with MR image acquisition processes. These distortions result from imperfections in the main magnetic field, nonlinear gradients, as well as field disturbances introduced by the imaged object. In this thesis, we quantified the effect of system related and patient-induced susceptibility geometric distortions on dose distributions for prostate as well as head and neck cancers. Methods to mitigate these distortions were also studied.

In Study I, mean worst system related residual distortions of 3.19, 2.52 and 2.08 mm at bandwidths (BW) of 122, 244 and 488 Hz/pixel up to a radial distance of 25 cm from a 3T PET/MR scanner was measured with a large field of view (FoV) phantom. Subsequently, we estimated maximum shifts of 5.8, 2.9 and 1.5 mm due to patient-induced susceptibility distortions. VMAT-optimized treatment plans initially performed on distorted CT (dCT) images and recalculated on real CT datasets resulted in a dose difference of less than 0.5%.

 The magnetic susceptibility differences at tissue-metallic,-air and -bone interfaces result in local B₀ magnetic field inhomogeneities. The distortion shifts caused by these field inhomogeneities can be reduced by shimming.  Study II aimed to investigate the use of shimming to improve the homogeneity of local  B₀ magnetic field which will be beneficial for radiotherapy applications. A shimming simulation based on spherical harmonics modeling was developed. The spinal cord, an organ at risk is surrounded by bone and in close proximity to the lungs may have high susceptibility differences. In this region, mean pixel shifts caused by local B₀ field inhomogeneities were reduced from 3.47±1.22 mm to 1.35±0.44 mm and 0.99±0.30 mm using first and second order shimming respectively. This was for a bandwidth of 122 Hz/pixel and an in-plane voxel size of 1×1 mm².  Also examined in Study II as in Study I was the dosimetric effect of geometric distortions on 21 Head and Neck cancer treatment plans. The dose difference in D₅₀ at the PTV between distorted CT and real CT plans was less than 1.0%.

In conclusion, the effect of MR geometric distortions on dose plans was small. Generally, we found patient-induced susceptibility distortions were larger compared with residual system distortions at all delineated structures except the external contour. This information will be relevant when setting margins for treatment volumes and organs at risk.  

The current practice of characterizing MR geometric distortions utilizing spatial accuracy phantoms alone may not be enough for an MR-only radiotherapy workflow. Therefore, measures to mitigate patient-induced susceptibility effects in clinical practice such as patient-specific correction algorithms are needed to complement existing distortion reduction methods such as high acquisition bandwidth and shimming.

Purpose: To investigate the effect of magnetic resonance system-and patient-induced susceptibility distortions from a 3T scanner on dose distributions for prostate cancers.

Methods and Materials: Combined displacement fields from the residual system and patient-induced susceptibility distortions were used to distort 17 prostate patient CT images. VMAT dose plans were initially optimized on distorted CT images and the plan parameters transferred to the original patient CT images to calculate a new dose distribution.

Results: Maximum residual mean distortions of 3.19 mm at a radial distance of 25 cm and maximum mean patient-induced susceptibility shifts of 5.8 mm were found using the lowest bandwidth of 122 Hz per pixel. There was a dose difference of <0.5% between distorted and undistorted treatment plans. The 90% confidence intervals of the mean difference between the dCT and CT treatment plans were all within an equivalence interval of (-0.5, 0.5) for all investigated plan quality measures.

Conclusions: Patient-induced susceptibility distortions at high field strengths in closed bore magnetic resonance scanners are larger than residual system distortions after using vendor-supplied 3-dimensional correction for the delineated regions studied. However, errors in dose due to disturbed patient outline and shifts caused by patient-induced susceptibility effects are below 0.5%.

Purpose: To evaluate the effect of magnetic resonance (MR) imaging (MRI) geometric distortions on head and neck radiation therapy treatment planning (RTP) for an MRI-only RTP. We also assessed the potential benefits of patient-specific shimming to reduce the magnitude of MR distortions for a 3-T scanner.

Methods and Materials: Using an in-house Matlab algorithm, shimming within entire imaging volumes and user-defined regions of interest were simulated. We deformed 21 patient computed tomography (CT) images with MR distortion fields (gradient nonlinearity and patient-induced susceptibility effects) to create distorted CT (dCT) images using bandwidths of 122 and 488 Hz/mm at 3 T. Field parameters from volumetric modulated arc therapy plans initially optimized on dCT data sets were transferred to CT data to compute a new plan. Both plans were compared to determine the impact of distortions on dose distributions.

Results: Shimming across entire patient volumes decreased the percentage of voxels with distortions of more than 2 mm from 15.4% to 2.0%. Using the user-defined region of interest (ROI) shimming strategy, (here the Planning target volume (PTV) was the chosen ROI volume) led to increased geometric for volumes outside the PTV, as such voxels within the spinal cord with geometric shifts above 2 mm increased from 11.5% to 32.3%. The worst phantom-measured residual system distortions after 3-dimensional gradient nonlinearity correction within a radial distance of 200 mm from the isocenter was 2.17 mm. For all patients, voxels with distortion shifts of more than 2 mm resulting from patient-induced susceptibility effects were 15.4% and 0.0% using bandwidths of 122 Hz/mm and 488 Hz/mm at 3 T. Dose differences between dCT and CT treatment plans in D-50 at the planning target volume were 0.4% +/- 0.6% and 0.3% +/- 0.5% at 122 and 488 Hz/mm, respectively.

Conclusions: The overall effect of MRI geometric distortions on data used for RTP was minimal. Shimming over entire imaging volumes decreased distortions, but user-defined subvolume shimming introduced significant errors in nearby organs and should probably be avoided.

Purpose: To investigate the effect of magnetic resonance system- and patient-induced susceptibility distortions from a 3T scanner on dose distributions for prostate cancers.

Methods and Materials: Combined displacement fields from the residual system and patient-induced susceptibility distortions were used to distort 17 prostate patient CT images. VMAT dose plans were initially optimized on distorted CT images and the plan parameters transferred to the original patient CT images to calculate a new dose distribution.

Results: Maximum residual mean distortions of 3.19 mm at a radial distance of 25 cm and maximum mean patient-induced susceptibility shifts of 5.8 mm were found using the lowest bandwidth of 122 Hz per pixel. There was a dose difference of <0.5% between distorted and undistorted treatment plans. The 90% confidence intervals of the mean difference between the dCT and CT treatment plans were all within an equivalence interval of (−0.5, 0.5) for all investigated plan quality measures.

Conclusions: Patient-induced susceptibility distortions at high field strengths in closed bore magnetic resonance scanners are larger than residual system distortions after using vendor-supplied 3-dimensional correction for the delineated regions studied. However, errors in dose due to disturbed patient outline and shifts caused by patient-induced susceptibility effects are below 0.5%.

Lithium formate has shown to be a material with properties suitable for electron paramagnetic resonance (EPR) dosimetry, among them up to 7 times higher sensitivity compared to alanine, which is a well-established EPR detector material for dose determinations in radiotherapy.

The aim of this thesis was to further investigate the properties of lithium formate and develop the dosimetry system towards applications in radiotherapy. The intrinsic efficiency for energies of relevance to brachytherapy and the signal stability were investigated. The dosimetry system was expanded to include a smaller dosimeter model, suitable for measurements in dose gradient regions. An individual sensitivity correction method was applied to the smaller dosimeters to be able to perform dose determinations with the same precision as for the larger ones. EPR dosimetry in general is time consuming and effort was spent to optimize the signal readout procedure regarding measurement time and measurement precision.

The system was applied in two clinical applications chosen for their high demands on the dosimetry system: 1) a dosimetry audit for external photon beam therapy and 2) dose verification measurements around a low energy HDR brachytherapy source.

The conclusions drawn from this thesis were: dose determinations can be performed with a standard uncertainty of 1.8-2.5% using both the original size dosimeters and the new developed smaller ones. The dosimetry system is robust and useful for applications when high measurement precision and accuracy is prioritized. It is a good candidate for dosimetry audits, both in external beam therapy and brachytherapy.

Clinical applications of electron paramagnetic resonance (EPR) dosimetry systems demand high accuracy causing time consuming analysis. The need for high spatial resolution dose measurements in regions with steep dose gradients demands small sized dosimeters. An optimization of the analysis was therefore needed to limit the time consumption. The aim of this work was to introduce a new smaller lithium formate dosimeter model (diameter reduced from standard diameter 4.5 mm to 3 mm and height from 4.8 mm to 3 mm). To compensate for reduced homogeneity in a batch of the smaller dosimeters, a method for individual sensitivity correction suitable for EPR dosimetry was tested. Sensitivity and repeatability was also tested for a standard EPR resonator and a super high Q (SHQE) one. The aim was also to optimize the performance of the dosimetry system for better efficiency regarding measurement time and precision. A systematic investigation of the relationship between measurement uncertainty and number of readouts per dosimeter was performed. The conclusions drawn from this work were that it is possible to decrease the dosimeter size with maintained measurement precision by using the SHQE resonator and introducing individual calibration factors for dosimeter batches. It was also shown that it is possible reduce the number of readouts per dosimeter without significantly decreasing the accuracy in measurements.

The aim of this work was to develop and test a remote end-to-end audit system using lithium formate EPR dosimeters. Four clinics were included in a pilot study, absorbed doses determined in the PTV agreed with TPS calculated doses within ±5% for 3D-CRT and ±7% (k=1) for IMRT/VMAT dose plans.

The aim of this study was to test two different solid state dosimetry systems for the purpose of end-to-end audits of radiotherapy volumetric modulated arc therapy (VMAT) technique; a lithium formate electron paramagnetic resonance system and a lithium fluoride thermoluminescent dosimetry system. As a complement to the solid state systems, ion chamber measurements were performed. A polystyrene phantom with a planning target volume (PTV) and an organ at risk (OAR) structure was scanned using CT. A VMAT dose plan was optimized to deliver 2 Gy to the target volume and to minimize the dose to the OAR. The different detectors were inserted into the phantom and the planned dose distribution was delivered. The measured doses were compared to the treatment planning system (TPS) calculated doses. Good agreement was found between the TPS calculated and the measured doses, well accepted for the dose determinations in remote dosimetry audits of VMAT treatment technique.

Interest in high dose rate (HDR) electronic brachytherapy operating at 50 kV is increasing. For quality assurance it is important to identify dosimetry systems that can measure the absorbed doses in absolute terms which is difficult in this energy region. In this work a comparison is made between two dosimetry systems, EPR lithium formate dosimeters and radiochromic EBT2 film.

Both types of dosimeters were irradiated simultaneously in a PMMA phantom using the Axxent EBS. Absorbed dose to water was determined at distances of 10 mm, 30 mm and 50 mm from the EBS. Results were traceable to different primary standards as regards to absorbed dose to water (EPR) and air kerma (EBT2). Monte Carlo simulations were used in absolute terms as a third estimate of absorbed dose to water.

Agreement within the estimated expanded (k = 2) uncertainties (5% (EPR), 7% (EBT2)) was found between the results at 30 mm and 50 mm from the x-ray source. The same result was obtained in 4 repetitions of irradiation, indicating high precision in the measurements with both systems. At all distances, agreement between EPR and Monte Carlo simulations was shown as was also the case for the film measurements at 30mm and 50mm. At 10mm the geometry for the film measurements caused too large uncertainty in measured values depending on the exact position (within sub-mm distances) of the EBS and the 10 mm film results were exculded from comparison.

This work has demonstrated good performance of the lithium formate EPR dosimetry system in accordance with earlier experiments at higher photon energies (¹⁹²Ir HDR brachytherapy). It was also highlighted that there might be issues regarding the energy dependence and intrinsic efficiency of the EBT2 film that need to be considered for measurements using low energy sources.

Advancements in generative artificial intelligence (AI) are setting the stage for transformative changes in medical imaging, particularly through the development of the Virtual Scanner. This innovative approach leverages resilient generative AI to synthesize radiological images, addressing critical challenges in the field such as data scarcity, patient exposure to radiation, and the limitations of current imaging technologies. By harnessing the power of Generative Adversarial Networks (GANs) and focusing on the resilience of these algorithms, the Virtual Scanner aims to enhance diagnostic accuracy, improve patient care, and fill gaps in multimodal datasets. Our research explores both unimodal and multimodal techniques, including GAN ensembles, latent augmentation, and advanced texture synthesis, to create robust and adaptable generative models. Through extensive experimentation and analysis, we demonstrate the potential of the Virtual Scanner to revolutionize medical diagnostics by providing a safer, more efficient, and comprehensive imaging solution. The implications of this work extend beyond immediate medical applications, offering insights into the development of AI technologies capable of navigating the complexities of real-world data.

Objective: [¹⁸F]FE-PE2I (FE-PE2I) is a new radiotracer for dopamine transporter (DAT) imaging with PET. The aim of this study was to evaluate the visual interpretation of FE-PE2I images for the diagnosis of idiopathic Parkinsonian syndrome (IPS). The inter-rater variability, sensitivity, specificity, and diagnostic accuracy for visual interpretation of striatal FE-PE2I compared to [¹²³I]FP-CIT (FP-CIT) single-photon emission computed tomography (SPECT) was evaluated.

Methods: Thirty patients with newly onset parkinsonism and 32 healthy controls with both an FE-PE2I and FP-CIT were included in the study. Four patients had normal DAT imaging, of which three did not fulfil the IPS criteria at the clinical reassessment after 2 years. Six raters evaluated the DAT images blinded to the clinical diagnosis, interpreting the image as being ‘normal’ or ‘pathological’, and assessed the degree of DAT-reduction in the caudate and putamen. The inter-rater agreement was assessed with intra-class correlation and Cronbach’s α. For calculation of sensitivity and specificity, DAT images were defined as correctly classified if categorized as normal or pathological by ≥4/6 raters.

Results: The overall agreement in visual evaluation of the FE-PE2I- and FP-CIT images was high for the IPS patients (α = 0.960 and 0.898, respectively), but lower in healthy controls (FE-PE2I: α = 0.693, FP-CIT: α = 0.657). Visual interpretation gave high sensitivity (both 0.96) but lower specificity (FE-PE2I: 0.86, FP-CIT: 0.63) with an accuracy of 90% for FE-PE2I and 77% for FP-CIT.

Conclusion: Visual evaluation of FE-PE2I PET imaging demonstrates high reliability and diagnostic accuracy for IPS.

Lung field segmentation is the first step towards the development of any computer aided diagnosis (CAD) system for interstitial lung diseases (ILD) observed in chest high resolution computed tomography (HRCT) images. If the segmentation is not done efficiently it will compromise the accuracy of CAD system. In this paper, a deep learning-based method is proposed to localize several interstitial lung disease patterns (ILD) in HRCT images without performing lung field segmentation. In this paper, localization of several ILD patterns is performed in image slice. The pretrained models of ZF and VGG networks were fine-tuned in order to localize ILD patterns using Faster R-CNN framework. The three most difficult ILD patterns consolidation, emphysema, and fibrosis have been used for this study and the accuracy of the method has been evaluated in terms of mean average precision (mAP) and free receiver operating characteristic (FROC) curve. The model achieved mAP value of 75% and 83% on ZF and VGG networks, respectively. The result obtained shows the effectiveness of the method in the localization of different ILD patterns.